{"caption": "", "dataset": "aistpp_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 161222, "index": 5, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00000.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/labels/gBR_sBM_cAll_d06_mBR2_ch02_retarget__seg0000_000000_000240.json", "segment_action": "a dancer steps side to side, swings arms, twists torso, bends knees, shifts weight rhythmically", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "groovy rhythmic", "sequence_name": "gBR_sBM_cAll_d06_mBR2_ch02_retarget__seg0000_000000_000240", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/g1/gBR_sBM_cAll_d06_mBR2_ch02_retarget__seg0000_000000_000240.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a groovy rhythmic sequence with side steps, arm swings, torso twists, and knee bends", "window_end": 240, "window_start": 180}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00000.npz", "sample_index": 0}
{"caption": "", "dataset": "aistpp_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 161227, "index": 10, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00001.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/labels/gBR_sBM_cAll_d06_mBR2_ch04_retarget__seg0000_000000_000240.json", "segment_action": "a dancer steps side to side, lifts knees, twists torso, and moves arms rhythmically", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "groovy coordinated", "sequence_name": "gBR_sBM_cAll_d06_mBR2_ch04_retarget__seg0000_000000_000240", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/g1/gBR_sBM_cAll_d06_mBR2_ch04_retarget__seg0000_000000_000240.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a groovy coordinated dance with side steps, knee lifts, torso twists, and rhythmic arm movements", "window_end": 150, "window_start": 90}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00001.npz", "sample_index": 1}
{"caption": "", "dataset": "aistpp_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 161233, "index": 16, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00002.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/labels/gBR_sBM_cAll_d06_mBR2_ch06_retarget__seg0000_000000_000240.json", "segment_action": "a dancer waves arms, bends knees, steps side to side, twists torso, and shifts weight", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "groovy energetic rhythmic", "sequence_name": "gBR_sBM_cAll_d06_mBR2_ch06_retarget__seg0000_000000_000240", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/g1/gBR_sBM_cAll_d06_mBR2_ch06_retarget__seg0000_000000_000240.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs an energetic groovy sequence with arm gestures, knee bends, and rhythmic steps", "window_end": 150, "window_start": 90}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00002.npz", "sample_index": 2}
{"caption": "", "dataset": "aistpp_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 161249, "index": 32, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00003.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/labels/gBR_sBM_cAll_d06_mBR3_ch05_retarget__seg0000_000000_000240.json", "segment_action": "a dancer bends torso, swings arms, steps side to side, and shifts weight rhythmically", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "bouncy rhythmic", "sequence_name": "gBR_sBM_cAll_d06_mBR3_ch05_retarget__seg0000_000000_000240", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/g1/gBR_sBM_cAll_d06_mBR3_ch05_retarget__seg0000_000000_000240.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a bouncy rhythmic sequence with torso bends, arm swings, and weight shifts", "window_end": 240, "window_start": 180}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00003.npz", "sample_index": 3}
{"caption": "", "dataset": "aistpp_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 161260, "index": 43, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00004.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/labels/gBR_sBM_cAll_d06_mBR4_ch03_retarget__seg0000_000000_000240.json", "segment_action": "a dancer steps side to side, bends knees, swings arms, twists torso rhythmically", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "groovy coordinated", "sequence_name": "gBR_sBM_cAll_d06_mBR4_ch03_retarget__seg0000_000000_000240", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/g1/gBR_sBM_cAll_d06_mBR4_ch03_retarget__seg0000_000000_000240.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a groovy coordinated dance with side steps, knee bends, and arm swings", "window_end": 150, "window_start": 90}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00004.npz", "sample_index": 4}
{"caption": "", "dataset": "aistpp_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 161273, "index": 56, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00005.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/labels/gBR_sFM_cAll_d06_mBR1_ch15_retarget__seg0000_000000_000240.json", "segment_action": "a dancer sweeps arms, bends knees, steps side to side, kneels, and twists torso", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "rhythmic coordinated", "sequence_name": "gBR_sFM_cAll_d06_mBR1_ch15_retarget__seg0000_000000_000240", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/g1/gBR_sFM_cAll_d06_mBR1_ch15_retarget__seg0000_000000_000240.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a rhythmic coordinated sequence with arm sweeps, knee bends, steps, and torso twists", "window_end": 240, "window_start": 180}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00005.npz", "sample_index": 5}
{"caption": "", "dataset": "aistpp_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 161283, "index": 66, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00006.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/labels/gBR_sFM_cAll_d06_mBR1_ch15_retarget__seg0006_000720_000960.json", "segment_action": "a dancer bends forward, raises arms, twists torso, and shifts weight while stepping rhythmically", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "expressive dynamic", "sequence_name": "gBR_sFM_cAll_d06_mBR1_ch15_retarget__seg0006_000720_000960", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/g1/gBR_sFM_cAll_d06_mBR1_ch15_retarget__seg0006_000720_000960.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs an expressive dynamic sequence with forward bends, arm raises, torso twists, and rhythmic steps", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00006.npz", "sample_index": 6}
{"caption": "", "dataset": "aistpp_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 161296, "index": 79, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00007.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/labels/gBR_sFM_cAll_d06_mBR2_ch16_retarget__seg0001_000120_000360.json", "segment_action": "a dancer lifts knees, swings arms, twists torso, and shifts weight rhythmically", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "energetic hip-hop", "sequence_name": "gBR_sFM_cAll_d06_mBR2_ch16_retarget__seg0001_000120_000360", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/g1/gBR_sFM_cAll_d06_mBR2_ch16_retarget__seg0001_000120_000360.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs an energetic hip-hop sequence with knee lifts, arm swings, and rhythmic steps", "window_end": 150, "window_start": 90}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00007.npz", "sample_index": 7}
{"caption": "", "dataset": "aistpp_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 161363, "index": 146, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00008.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/labels/gHO_sFM_cAll_d21_mHO0_ch15_retarget__seg0002_000240_000480.json", "segment_action": "a dancer bends knees, swings arms, twists torso, shifts weight, and spins rhythmically", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "groovy rhythmic", "sequence_name": "gHO_sFM_cAll_d21_mHO0_ch15_retarget__seg0002_000240_000480", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/g1/gHO_sFM_cAll_d21_mHO0_ch15_retarget__seg0002_000240_000480.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a groovy rhythmic dance with knee bends, arm swings, spins, and weight shifts", "window_end": 240, "window_start": 180}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00008.npz", "sample_index": 8}
{"caption": "", "dataset": "aistpp_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 161375, "index": 158, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00009.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/labels/gHO_sFM_cAll_d21_mHO0_ch15_retarget__seg0006_000720_000960.json", "segment_action": "a dancer bends knees, swings arms, shifts weight, spins, and steps rhythmically", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "funky coordinated energetic", "sequence_name": "gHO_sFM_cAll_d21_mHO0_ch15_retarget__seg0006_000720_000960", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/g1/gHO_sFM_cAll_d21_mHO0_ch15_retarget__seg0006_000720_000960.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a funky coordinated dance with rhythmic knee bends, arm swings, weight shifts, and spins", "window_end": 240, "window_start": 180}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00009.npz", "sample_index": 9}
{"caption": "", "dataset": "aistpp_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 161393, "index": 176, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00010.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/labels/gHO_sFM_cAll_d21_mHO1_ch16_retarget__seg0005_000600_000840.json", "segment_action": "a dancer spins, moves arms fluidly, bends knees, and shifts weight with side steps", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "groovy rhythmic", "sequence_name": "gHO_sFM_cAll_d21_mHO1_ch16_retarget__seg0005_000600_000840", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/g1/gHO_sFM_cAll_d21_mHO1_ch16_retarget__seg0005_000600_000840.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a groovy rhythmic dance with spins, arm gestures, and coordinated steps", "window_end": 240, "window_start": 180}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00010.npz", "sample_index": 10}
{"caption": "", "dataset": "aistpp_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 161463, "index": 246, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00011.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/labels/gHO_sFM_cAll_d21_mHO5_ch20_retarget__seg0002_000240_000480.json", "segment_action": "a dancer bends knees, steps side to side, swings arms, twists torso, shifts weight rhythmically", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "bouncy rhythmic", "sequence_name": "gHO_sFM_cAll_d21_mHO5_ch20_retarget__seg0002_000240_000480", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/g1/gHO_sFM_cAll_d21_mHO5_ch20_retarget__seg0002_000240_000480.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a bouncy rhythmic sequence with knee bends, side steps, arm swings, and torso twists", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00011.npz", "sample_index": 11}
{"caption": "", "dataset": "aistpp_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 161558, "index": 341, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00012.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/labels/gJB_sFM_cAll_d09_mJB5_ch20_retarget__seg0002_000240_000480.json", "segment_action": "a dancer moves arms fluidly, bends torso, spins, steps side to side, and shifts weight rhythmically", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "expressive graceful", "sequence_name": "gJB_sFM_cAll_d09_mJB5_ch20_retarget__seg0002_000240_000480", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/g1/gJB_sFM_cAll_d09_mJB5_ch20_retarget__seg0002_000240_000480.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs an expressive graceful sequence with fluid arm movements, torso bends, spins, and rhythmic steps", "window_end": 240, "window_start": 180}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00012.npz", "sample_index": 12}
{"caption": "", "dataset": "aistpp_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 161563, "index": 346, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00013.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/labels/gJS_sBM_cAll_d03_mJS2_ch02_retarget__seg0000_000000_000240.json", "segment_action": "a dancer takes small side steps, shifts weight, and moves hips gently", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "slow rhythmic", "sequence_name": "gJS_sBM_cAll_d03_mJS2_ch02_retarget__seg0000_000000_000240", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/g1/gJS_sBM_cAll_d03_mJS2_ch02_retarget__seg0000_000000_000240.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs slow rhythmic side steps with subtle hip shifts", "window_end": 150, "window_start": 90}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00013.npz", "sample_index": 13}
{"caption": "", "dataset": "aistpp_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 161741, "index": 524, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00014.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/labels/gJS_sFM_cAll_d03_mJS5_ch14_retarget__seg0001_000120_000360.json", "segment_action": "a dancer bends torso, steps rhythmically, raises arms, and shifts weight with coordinated movements", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "playful expressive", "sequence_name": "gJS_sFM_cAll_d03_mJS5_ch14_retarget__seg0001_000120_000360", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/g1/gJS_sFM_cAll_d03_mJS5_ch14_retarget__seg0001_000120_000360.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a playful expressive sequence with torso bends, arm gestures, and rhythmic steps", "window_end": 240, "window_start": 180}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00014.npz", "sample_index": 14}
{"caption": "", "dataset": "aistpp_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 161760, "index": 543, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00015.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/labels/gKR_sBM_cAll_d30_mKR2_ch03_retarget__seg0000_000000_000240.json", "segment_action": "a dancer raises arms, twists torso, steps side to side with coordinated arm gestures", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "groovy rhythmic", "sequence_name": "gKR_sBM_cAll_d30_mKR2_ch03_retarget__seg0000_000000_000240", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/g1/gKR_sBM_cAll_d30_mKR2_ch03_retarget__seg0000_000000_000240.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a groovy rhythmic sequence with arm gestures, torso twists, and side steps", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00015.npz", "sample_index": 15}
{"caption": "", "dataset": "aistpp_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 161809, "index": 592, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00016.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/labels/gKR_sBM_cAll_d30_mKR3_ch09_retarget__seg0000_000000_000240.json", "segment_action": "a dancer twists torso, swings arms, sways hips, and shifts weight in rhythmic motions", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "funky rhythmic", "sequence_name": "gKR_sBM_cAll_d30_mKR3_ch09_retarget__seg0000_000000_000240", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/g1/gKR_sBM_cAll_d30_mKR3_ch09_retarget__seg0000_000000_000240.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a funky rhythmic sequence with torso twists, arm swings, hip sways, and weight shifts", "window_end": 150, "window_start": 90}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00016.npz", "sample_index": 16}
{"caption": "", "dataset": "aistpp_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 161815, "index": 598, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00017.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/labels/gKR_sBM_cAll_d30_mKR4_ch01_retarget__seg0000_000000_000240.json", "segment_action": "a dancer bends knees, swings arms alternately, shifts weight, and tilts torso rhythmically", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "bouncy rhythmic", "sequence_name": "gKR_sBM_cAll_d30_mKR4_ch01_retarget__seg0000_000000_000240", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/g1/gKR_sBM_cAll_d30_mKR4_ch01_retarget__seg0000_000000_000240.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a bouncy rhythmic sequence with knee bends, arm swings, and weight shifts", "window_end": 150, "window_start": 90}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00017.npz", "sample_index": 17}
{"caption": "", "dataset": "aistpp_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 161991, "index": 774, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00018.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/labels/gLH_sBM_cAll_d18_mLH2_ch03_retarget__seg0000_000000_000240.json", "segment_action": "a dancer steps side to side, twists torso, and moves arms in sync with weight shifts", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "groovy rhythmic", "sequence_name": "gLH_sBM_cAll_d18_mLH2_ch03_retarget__seg0000_000000_000240", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/g1/gLH_sBM_cAll_d18_mLH2_ch03_retarget__seg0000_000000_000240.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a groovy rhythmic sequence with side steps, torso twists, and arm movements", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00018.npz", "sample_index": 18}
{"caption": "", "dataset": "aistpp_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 162015, "index": 798, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00019.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/labels/gLH_sBM_cAll_d18_mLH3_ch01_retarget__seg0000_000000_000240.json", "segment_action": "a dancer makes arm gestures, twists torso, steps side to side, shifts weight rhythmically", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "groovy rhythmic", "sequence_name": "gLH_sBM_cAll_d18_mLH3_ch01_retarget__seg0000_000000_000240", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/g1/gLH_sBM_cAll_d18_mLH3_ch01_retarget__seg0000_000000_000240.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a groovy rhythmic sequence with arm gestures, torso twists, and side steps", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00019.npz", "sample_index": 19}
{"caption": "", "dataset": "aistpp_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 162053, "index": 836, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00020.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/labels/gLH_sBM_cAll_d18_mLH4_ch03_retarget__seg0000_000000_000240.json", "segment_action": "a dancer steps side to side, twists torso, swings arms, and shifts weight with rhythmic body leans", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "groovy rhythmic", "sequence_name": "gLH_sBM_cAll_d18_mLH4_ch03_retarget__seg0000_000000_000240", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/g1/gLH_sBM_cAll_d18_mLH4_ch03_retarget__seg0000_000000_000240.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a groovy rhythmic sequence with side steps, torso twists, arm swings, and body leans", "window_end": 240, "window_start": 180}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00020.npz", "sample_index": 20}
{"caption": "", "dataset": "aistpp_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 162168, "index": 951, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00021.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/labels/gLH_sFM_cAll_d18_mLH3_ch18_retarget__seg0000_000000_000240.json", "segment_action": "a dancer moves arms dynamically, sways torso, steps side to side, shifts weight rhythmically", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "groovy rhythmic", "sequence_name": "gLH_sFM_cAll_d18_mLH3_ch18_retarget__seg0000_000000_000240", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/g1/gLH_sFM_cAll_d18_mLH3_ch18_retarget__seg0000_000000_000240.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a groovy rhythmic dance with arm gestures, torso sways, and side steps", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00021.npz", "sample_index": 21}
{"caption": "", "dataset": "aistpp_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 162200, "index": 983, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00022.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/labels/gLH_sFM_cAll_d18_mLH4_ch19_retarget__seg0003_000360_000600.json", "segment_action": "a dancer swings arms, twists torso, steps side to side, and sways hips in coordinated motions", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "groovy rhythmic", "sequence_name": "gLH_sFM_cAll_d18_mLH4_ch19_retarget__seg0003_000360_000600", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/g1/gLH_sFM_cAll_d18_mLH4_ch19_retarget__seg0003_000360_000600.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a groovy rhythmic dance with arm swings, torso twists, side steps, and hip sways", "window_end": 240, "window_start": 180}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00022.npz", "sample_index": 22}
{"caption": "", "dataset": "aistpp_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 162209, "index": 992, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00023.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/labels/gLH_sFM_cAll_d18_mLH4_ch19_retarget__seg0006_000720_000960.json", "segment_action": "a dancer makes arm gestures, twists torso, bends knees, steps side to side, shifts weight", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "playful dynamic", "sequence_name": "gLH_sFM_cAll_d18_mLH4_ch19_retarget__seg0006_000720_000960", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/g1/gLH_sFM_cAll_d18_mLH4_ch19_retarget__seg0006_000720_000960.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a playful dynamic dance with arm gestures, torso twists, knee bends, and rhythmic steps", "window_end": 240, "window_start": 180}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00023.npz", "sample_index": 23}
{"caption": "", "dataset": "aistpp_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 162260, "index": 1043, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00024.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/labels/gLO_sBM_cAll_d15_mLO3_ch01_retarget__seg0000_000000_000240.json", "segment_action": "a dancer bends knees, leans torso, shifts weight, extends one arm, and repeats", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "deliberate slow", "sequence_name": "gLO_sBM_cAll_d15_mLO3_ch01_retarget__seg0000_000000_000240", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/g1/gLO_sBM_cAll_d15_mLO3_ch01_retarget__seg0000_000000_000240.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a deliberate slow sequence with knee bends, torso leans, and weight shifts", "window_end": 240, "window_start": 180}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00024.npz", "sample_index": 24}
{"caption": "", "dataset": "aistpp_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 162280, "index": 1063, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00025.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/labels/gLO_sBM_cAll_d15_mLO3_ch08_retarget__seg0000_000000_000240.json", "segment_action": "a dancer steps side to side, flicks arms, bends torso, and shifts hips rhythmically", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "groovy rhythmic", "sequence_name": "gLO_sBM_cAll_d15_mLO3_ch08_retarget__seg0000_000000_000240", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/g1/gLO_sBM_cAll_d15_mLO3_ch08_retarget__seg0000_000000_000240.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a groovy rhythmic sequence with side steps, arm flicks, torso bends, and hip shifts", "window_end": 150, "window_start": 90}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00025.npz", "sample_index": 25}
{"caption": "", "dataset": "aistpp_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 162286, "index": 1069, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00026.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/labels/gLO_sBM_cAll_d15_mLO3_ch10_retarget__seg0000_000000_000240.json", "segment_action": "a dancer bends torso, swings arms, lifts knees, and shifts weight rhythmically", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "funky energetic", "sequence_name": "gLO_sBM_cAll_d15_mLO3_ch10_retarget__seg0000_000000_000240", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/g1/gLO_sBM_cAll_d15_mLO3_ch10_retarget__seg0000_000000_000240.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a funky energetic dance with torso bends, arm swings, knee lifts, and rhythmic weight shifts", "window_end": 150, "window_start": 90}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00026.npz", "sample_index": 26}
{"caption": "", "dataset": "aistpp_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 162306, "index": 1089, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00027.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/labels/gLO_sBM_cAll_d15_mLO4_ch08_retarget__seg0000_000000_000240.json", "segment_action": "a dancer bends knees, taps feet rhythmically, sways torso, and moves arms in sync", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "bouncy rhythmic", "sequence_name": "gLO_sBM_cAll_d15_mLO4_ch08_retarget__seg0000_000000_000240", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/g1/gLO_sBM_cAll_d15_mLO4_ch08_retarget__seg0000_000000_000240.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a bouncy rhythmic sequence with knee bends, foot taps, and coordinated arm movements", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00027.npz", "sample_index": 27}
{"caption": "", "dataset": "aistpp_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 162313, "index": 1096, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00028.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/labels/gLO_sBM_cAll_d15_mLO4_ch10_retarget__seg0000_000000_000240.json", "segment_action": "a dancer bends knees, swings arms, twists torso, and shifts weight rhythmically", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "energetic hip-hop", "sequence_name": "gLO_sBM_cAll_d15_mLO4_ch10_retarget__seg0000_000000_000240", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/g1/gLO_sBM_cAll_d15_mLO4_ch10_retarget__seg0000_000000_000240.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs an energetic hip-hop sequence with knee bends, arm swings, torso twists, and rhythmic footwork", "window_end": 150, "window_start": 90}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00028.npz", "sample_index": 28}
{"caption": "", "dataset": "aistpp_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 162401, "index": 1184, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00029.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/labels/gLO_sFM_cAll_d15_mLO3_ch18_retarget__seg0001_000120_000360.json", "segment_action": "a dancer sweeps arms, twists torso, bends knees, steps side to side, shifts weight rhythmically", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "groovy rhythmic", "sequence_name": "gLO_sFM_cAll_d15_mLO3_ch18_retarget__seg0001_000120_000360", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/g1/gLO_sFM_cAll_d15_mLO3_ch18_retarget__seg0001_000120_000360.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a groovy rhythmic dance with arm sweeps, torso twists, knee bends, and side steps", "window_end": 240, "window_start": 180}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00029.npz", "sample_index": 29}
{"caption": "", "dataset": "aistpp_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 162453, "index": 1236, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00030.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/labels/gLO_sFM_cAll_d15_mLO4_ch21_retarget__seg0005_000600_000840.json", "segment_action": "a dancer waves arms, twists torso, sways hips, bends knees, and shifts weight", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "expressive coordinated", "sequence_name": "gLO_sFM_cAll_d15_mLO4_ch21_retarget__seg0005_000600_000840", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/g1/gLO_sFM_cAll_d15_mLO4_ch21_retarget__seg0005_000600_000840.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs an expressive coordinated dance with arm gestures, torso twists, and hip sways", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00030.npz", "sample_index": 30}
{"caption": "", "dataset": "aistpp_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 162483, "index": 1266, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00031.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/labels/gMH_sBM_cAll_d24_mMH2_ch03_retarget__seg0000_000000_000240.json", "segment_action": "a dancer bends knees, swings arms, twists torso, and steps side to side with weight shifts", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "groovy rhythmic", "sequence_name": "gMH_sBM_cAll_d24_mMH2_ch03_retarget__seg0000_000000_000240", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/g1/gMH_sBM_cAll_d24_mMH2_ch03_retarget__seg0000_000000_000240.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a groovy rhythmic dance with knee bends, arm swings, and side steps", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00031.npz", "sample_index": 31}
{"caption": "", "dataset": "aistpp_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 162533, "index": 1316, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00032.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/labels/gMH_sBM_cAll_d24_mMH3_ch09_retarget__seg0000_000000_000240.json", "segment_action": "a dancer bends knees, steps in place, twists torso, and moves arms rhythmically forward and back", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "bouncy rhythmic", "sequence_name": "gMH_sBM_cAll_d24_mMH3_ch09_retarget__seg0000_000000_000240", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/g1/gMH_sBM_cAll_d24_mMH3_ch09_retarget__seg0000_000000_000240.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a bouncy rhythmic sequence with knee bends, torso twists, and arm movements", "window_end": 240, "window_start": 180}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00032.npz", "sample_index": 32}
{"caption": "", "dataset": "aistpp_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 162548, "index": 1331, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00033.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/labels/gMH_sBM_cAll_d24_mMH4_ch04_retarget__seg0000_000000_000240.json", "segment_action": "a dancer swings arms alternately, bends knees, shifts weight side to side, and sways torso", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "groovy rhythmic", "sequence_name": "gMH_sBM_cAll_d24_mMH4_ch04_retarget__seg0000_000000_000240", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/g1/gMH_sBM_cAll_d24_mMH4_ch04_retarget__seg0000_000000_000240.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a groovy rhythmic sequence with arm swings, knee bends, and side weight shifts", "window_end": 240, "window_start": 180}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00033.npz", "sample_index": 33}
{"caption": "", "dataset": "aistpp_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 162643, "index": 1426, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00034.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/labels/gMH_sFM_cAll_d24_mMH2_ch17_retarget__seg0006_000720_000960.json", "segment_action": "a dancer twists torso, sweeps arms, bends knees, steps side to side, shifts weight dynamically", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "groovy rhythmic", "sequence_name": "gMH_sFM_cAll_d24_mMH2_ch17_retarget__seg0006_000720_000960", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/g1/gMH_sFM_cAll_d24_mMH2_ch17_retarget__seg0006_000720_000960.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a groovy rhythmic sequence with torso twists, arm sweeps, and dynamic footwork", "window_end": 150, "window_start": 90}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00034.npz", "sample_index": 34}
{"caption": "", "dataset": "aistpp_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 162653, "index": 1436, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00035.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/labels/gMH_sFM_cAll_d24_mMH2_ch21_retarget__seg0001_000120_000360.json", "segment_action": "a dancer sweeps arms, twists torso, bends knees, spins, and shifts weight rhythmically", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "expressive dynamic", "sequence_name": "gMH_sFM_cAll_d24_mMH2_ch21_retarget__seg0001_000120_000360", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/g1/gMH_sFM_cAll_d24_mMH2_ch21_retarget__seg0001_000120_000360.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs an expressive dynamic sequence with arm sweeps, spins, torso twists, and knee bends", "window_end": 240, "window_start": 180}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00035.npz", "sample_index": 35}
{"caption": "", "dataset": "aistpp_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 162722, "index": 1505, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00036.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/labels/gMH_sFM_cAll_d24_mMH5_ch20_retarget__seg0002_000240_000480.json", "segment_action": "a dancer waves arms, twists torso, bends knees, steps side to side with rhythmic shifts", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "energetic groovy hip-hop", "sequence_name": "gMH_sFM_cAll_d24_mMH5_ch20_retarget__seg0002_000240_000480", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/g1/gMH_sFM_cAll_d24_mMH5_ch20_retarget__seg0002_000240_000480.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs an energetic groovy hip-hop sequence with arm gestures, torso twists, and rhythmic footwork", "window_end": 240, "window_start": 180}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00036.npz", "sample_index": 36}
{"caption": "", "dataset": "aistpp_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 162802, "index": 1585, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00037.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/labels/gPO_sBM_cAll_d12_mPO4_ch04_retarget__seg0000_000000_000240.json", "segment_action": "a dancer turns torso side to side, holds arms outward, and shifts weight between feet", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "smooth rhythmic", "sequence_name": "gPO_sBM_cAll_d12_mPO4_ch04_retarget__seg0000_000000_000240", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/g1/gPO_sBM_cAll_d12_mPO4_ch04_retarget__seg0000_000000_000240.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a smooth rhythmic sequence with torso turns, extended arms, and weight shifts", "window_end": 150, "window_start": 90}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00037.npz", "sample_index": 37}
{"caption": "", "dataset": "aistpp_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 162816, "index": 1599, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00038.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/labels/gPO_sBM_cAll_d12_mPO4_ch09_retarget__seg0000_000000_000240.json", "segment_action": "a dancer steps side to side, bends knees rhythmically, and holds arms in a steady stance", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "rhythmic bouncy", "sequence_name": "gPO_sBM_cAll_d12_mPO4_ch09_retarget__seg0000_000000_000240", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/g1/gPO_sBM_cAll_d12_mPO4_ch09_retarget__seg0000_000000_000240.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a rhythmic bouncy sequence with side steps, knee bends, and steady arm positioning", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00038.npz", "sample_index": 38}
{"caption": "", "dataset": "aistpp_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 162867, "index": 1650, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00039.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/labels/gPO_sFM_cAll_d12_mPO1_ch16_retarget__seg0005_000600_000840.json", "segment_action": "a dancer waves arms, sways hips, bends knees, shifts weight, and makes dynamic upper-body moves", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "playful rhythmic", "sequence_name": "gPO_sFM_cAll_d12_mPO1_ch16_retarget__seg0005_000600_000840", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/g1/gPO_sFM_cAll_d12_mPO1_ch16_retarget__seg0005_000600_000840.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a playful rhythmic sequence with arm gestures, hip sways, and dynamic weight shifts", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00039.npz", "sample_index": 39}
{"caption": "", "dataset": "aistpp_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 162884, "index": 1667, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00040.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/labels/gPO_sFM_cAll_d12_mPO2_ch17_retarget__seg0001_000120_000360.json", "segment_action": "a dancer makes arm gestures, lifts knees, steps side to side, sways torso, and shifts weight", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "playful rhythmic", "sequence_name": "gPO_sFM_cAll_d12_mPO2_ch17_retarget__seg0001_000120_000360", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/g1/gPO_sFM_cAll_d12_mPO2_ch17_retarget__seg0001_000120_000360.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a playful rhythmic dance with arm gestures, knee lifts, side steps, and body sways", "window_end": 240, "window_start": 180}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00040.npz", "sample_index": 40}
{"caption": "", "dataset": "aistpp_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 162901, "index": 1684, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00041.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/labels/gPO_sFM_cAll_d12_mPO2_ch17_retarget__seg0007_000840_001080.json", "segment_action": "a dancer spins, bends knees, uses arm gestures, shifts weight, twists torso, and raises an arm", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "playful rhythmic", "sequence_name": "gPO_sFM_cAll_d12_mPO2_ch17_retarget__seg0007_000840_001080", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/g1/gPO_sFM_cAll_d12_mPO2_ch17_retarget__seg0007_000840_001080.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a playful rhythmic dance with spins, knee bends, arm gestures, and weight shifts", "window_end": 150, "window_start": 90}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00041.npz", "sample_index": 41}
{"caption": "", "dataset": "aistpp_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 162920, "index": 1703, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00042.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/labels/gPO_sFM_cAll_d12_mPO3_ch18_retarget__seg0005_000600_000840.json", "segment_action": "a dancer bends knees, twists torso, moves arms expressively, and shifts weight while stepping side to side", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "playful rhythmic", "sequence_name": "gPO_sFM_cAll_d12_mPO3_ch18_retarget__seg0005_000600_000840", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/g1/gPO_sFM_cAll_d12_mPO3_ch18_retarget__seg0005_000600_000840.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a playful rhythmic sequence with knee bends, arm gestures, torso twists, and side steps", "window_end": 240, "window_start": 180}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00042.npz", "sample_index": 42}
{"caption": "", "dataset": "aistpp_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 162997, "index": 1780, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00043.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/labels/gWA_sBM_cAll_d27_mWA2_ch06_retarget__seg0000_000000_000240.json", "segment_action": "a dancer swings arms, twists torso, steps side to side, shifts weight, and executes full turns", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "groovy rhythmic", "sequence_name": "gWA_sBM_cAll_d27_mWA2_ch06_retarget__seg0000_000000_000240", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/g1/gWA_sBM_cAll_d27_mWA2_ch06_retarget__seg0000_000000_000240.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a groovy rhythmic dance with arm swings, torso twists, side steps, and turns", "window_end": 150, "window_start": 90}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00043.npz", "sample_index": 43}
{"caption": "", "dataset": "aistpp_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 163048, "index": 1831, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00044.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/labels/gWA_sBM_cAll_d27_mWA4_ch07_retarget__seg0000_000000_000240.json", "segment_action": "a dancer raises arms, twists torso, steps side to side, makes rhythmic arm gestures, and shifts weight", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "playful coordinated", "sequence_name": "gWA_sBM_cAll_d27_mWA4_ch07_retarget__seg0000_000000_000240", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/g1/gWA_sBM_cAll_d27_mWA4_ch07_retarget__seg0000_000000_000240.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a playful coordinated dance with arm gestures, torso twists, and rhythmic weight shifts", "window_end": 150, "window_start": 90}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00044.npz", "sample_index": 44}
{"caption": "", "dataset": "aistpp_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 163113, "index": 1896, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00045.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/labels/gWA_sFM_cAll_d27_mWA2_ch17_retarget__seg0005_000600_000840.json", "segment_action": "a dancer sways hips, waves arms, twists torso, leans back, and shifts feet rhythmically", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "rhythmic coordinated", "sequence_name": "gWA_sFM_cAll_d27_mWA2_ch17_retarget__seg0005_000600_000840", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/g1/gWA_sFM_cAll_d27_mWA2_ch17_retarget__seg0005_000600_000840.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a rhythmic coordinated sequence with hip sways, arm gestures, and body leans", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00045.npz", "sample_index": 45}
{"caption": "", "dataset": "aistpp_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 163172, "index": 1955, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00046.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/labels/gWA_sFM_cAll_d27_mWA4_ch19_retarget__seg0002_000240_000480.json", "segment_action": "a dancer waves arms, twists torso, steps side to side, spins, and bends knees rhythmically", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "expressive coordinated", "sequence_name": "gWA_sFM_cAll_d27_mWA4_ch19_retarget__seg0002_000240_000480", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/aistpp/g1/gWA_sFM_cAll_d27_mWA4_ch19_retarget__seg0002_000240_000480.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs an expressive coordinated dance with arm gestures, torso twists, steps, and spins", "window_end": 240, "window_start": 180}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00046.npz", "sample_index": 46}
{"caption": "step forward with alternating legs, lower arms to the sides, and adjust stance to stand upright facing forward steadily; style: casual", "dataset": "amass_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 440, "index": 440, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00047.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/labels/BMLhandball/S07_Expert/Trial_upper_right_023_poses_retarget__seg0003__part0000.json", "segment_action": "step forward with alternating legs, lower arms to the sides, and adjust stance to stand upright facing forward steadily", "segment_frame_end": 174, "segment_frame_start": 0, "segment_index": 0, "segment_style": "casual", "sequence_name": "Trial_upper_right_023_poses_retarget__seg0003__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/g1/BMLhandball/S07_Expert/Trial_upper_right_023_poses_retarget__seg0003__part0000.npz", "split": "test", "text_path": null, "video_summary": "step forward with alternating legs, lower arms to the sides, and adjust stance to stand upright facing forward steadily; style: casual", "window_end": 174, "window_start": 114}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00047.npz", "sample_index": 47}
{"caption": "run forward with arms bent and clasped in front, body leaning forward; style: dynamic", "dataset": "amass_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 442, "index": 442, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00048.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/labels/BMLhandball/S07_Expert/Trial_upper_right_left_004_poses_retarget__seg0000__part0000.json", "segment_action": "run forward with arms bent and clasped in front, body leaning forward", "segment_frame_end": 84, "segment_frame_start": 0, "segment_index": 0, "segment_style": "dynamic", "sequence_name": "Trial_upper_right_left_004_poses_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/g1/BMLhandball/S07_Expert/Trial_upper_right_left_004_poses_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "run forward with arms bent and clasped in front, body leaning forward; style: dynamic", "window_end": 72, "window_start": 12}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00048.npz", "sample_index": 48}
{"caption": "walk backward, with arms initially held near torso then gradually lowering to sides; style: controlled", "dataset": "amass_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 470, "index": 470, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00049.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/labels/BMLhandball/S08_Novice/Trial_upper_left_117_poses_retarget__seg0003__part0000.json", "segment_action": "walk backward, with arms initially held near torso then gradually lowering to sides", "segment_frame_end": 164, "segment_frame_start": 0, "segment_index": 0, "segment_style": "controlled", "sequence_name": "Trial_upper_left_117_poses_retarget__seg0003__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/g1/BMLhandball/S08_Novice/Trial_upper_left_117_poses_retarget__seg0003__part0000.npz", "split": "test", "text_path": null, "video_summary": "walk backward, with arms initially held near torso then gradually lowering to sides; style: controlled", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00049.npz", "sample_index": 49}
{"caption": "rise from crouch, rotate torso away from camera, raise arms outward, lean back while shifting body weight; style: dynamic", "dataset": "amass_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 522, "index": 522, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00050.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/labels/BMLhandball/S09_Novice/Trial_upper_left_200_poses_retarget__seg0001__part0000.json", "segment_action": "rise from crouch, rotate torso away from camera, raise arms outward, lean back while shifting body weight", "segment_frame_end": 135, "segment_frame_start": 0, "segment_index": 0, "segment_style": "dynamic", "sequence_name": "Trial_upper_left_200_poses_retarget__seg0001__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/g1/BMLhandball/S09_Novice/Trial_upper_left_200_poses_retarget__seg0001__part0000.npz", "split": "test", "text_path": null, "video_summary": "rise from crouch, rotate torso away from camera, raise arms outward, lean back while shifting body weight; style: dynamic", "window_end": 135, "window_start": 75}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00050.npz", "sample_index": 50}
{"caption": "land on feet and continue spinning with bent knees; style: controlled", "dataset": "amass_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 717, "index": 717, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00051.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/labels/BMLmovi/Subject_24_F_MoSh/Subject_24_F_16_poses_retarget__seg0002__part0000.json", "segment_action": "land on feet and continue spinning with bent knees", "segment_frame_end": 69, "segment_frame_start": 0, "segment_index": 0, "segment_style": "controlled", "sequence_name": "Subject_24_F_16_poses_retarget__seg0002__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/g1/BMLmovi/Subject_24_F_MoSh/Subject_24_F_16_poses_retarget__seg0002__part0000.npz", "split": "test", "text_path": null, "video_summary": "land on feet and continue spinning with bent knees; style: controlled", "window_end": 69, "window_start": 9}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00051.npz", "sample_index": 51}
{"caption": "shift body weight, rotate torso, lower body gradually into a seated position; style: controlled", "dataset": "amass_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 780, "index": 780, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00052.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/labels/BMLmovi/Subject_34_F_MoSh/Subject_34_F_16_poses_retarget__seg0000__part0000.json", "segment_action": "shift body weight, rotate torso, lower body gradually into a seated position", "segment_frame_end": 84, "segment_frame_start": 0, "segment_index": 0, "segment_style": "controlled", "sequence_name": "Subject_34_F_16_poses_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/g1/BMLmovi/Subject_34_F_MoSh/Subject_34_F_16_poses_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "shift body weight, rotate torso, lower body gradually into a seated position; style: controlled", "window_end": 72, "window_start": 12}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00052.npz", "sample_index": 52}
{"caption": "move both arms in repetitive forward circular motions while slightly bending knees; style: smooth", "dataset": "amass_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 861, "index": 861, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00053.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/labels/BMLmovi/Subject_44_F_MoSh/Subject_44_F_17_poses_retarget__seg0001__part0000.json", "segment_action": "move both arms in repetitive forward circular motions while slightly bending knees", "segment_frame_end": 126, "segment_frame_start": 0, "segment_index": 0, "segment_style": "smooth", "sequence_name": "Subject_44_F_17_poses_retarget__seg0001__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/g1/BMLmovi/Subject_44_F_MoSh/Subject_44_F_17_poses_retarget__seg0001__part0000.npz", "split": "test", "text_path": null, "video_summary": "move both arms in repetitive forward circular motions while slightly bending knees; style: smooth", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00053.npz", "sample_index": 53}
{"caption": "alternate lifting legs while swinging arms in circular motions repeatedly; style: rhythmic", "dataset": "amass_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 1007, "index": 1007, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00054.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/labels/BMLmovi/Subject_60_F_MoSh/Subject_60_F_15_poses_retarget__seg0001__part0000.json", "segment_action": "alternate lifting legs while swinging arms in circular motions repeatedly", "segment_frame_end": 105, "segment_frame_start": 0, "segment_index": 0, "segment_style": "rhythmic", "sequence_name": "Subject_60_F_15_poses_retarget__seg0001__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/g1/BMLmovi/Subject_60_F_MoSh/Subject_60_F_15_poses_retarget__seg0001__part0000.npz", "split": "test", "text_path": null, "video_summary": "alternate lifting legs while swinging arms in circular motions repeatedly; style: rhythmic", "window_end": 105, "window_start": 45}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00054.npz", "sample_index": 54}
{"caption": "bend torso forward, bend knees, hold crouched stance with arms extended downward; style: steady", "dataset": "amass_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 1480, "index": 1480, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00055.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/labels/BioMotionLab_NTroje/rub021/0018_lifting_light2_poses_retarget__seg0001__part0000.json", "segment_action": "bend torso forward, bend knees, hold crouched stance with arms extended downward", "segment_frame_end": 75, "segment_frame_start": 0, "segment_index": 0, "segment_style": "steady", "sequence_name": "0018_lifting_light2_poses_retarget__seg0001__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/g1/BioMotionLab_NTroje/rub021/0018_lifting_light2_poses_retarget__seg0001__part0000.npz", "split": "test", "text_path": null, "video_summary": "bend torso forward, bend knees, hold crouched stance with arms extended downward; style: steady", "window_end": 68, "window_start": 8}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00055.npz", "sample_index": 55}
{"caption": "cross arms in front of body, then extend them outward alternately, repeat motion; style: fluid", "dataset": "amass_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 1926, "index": 1926, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00056.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/labels/BioMotionLab_NTroje/rub055/0031_rom_poses_retarget__seg0001__part0001.json", "segment_action": "cross arms in front of body, then extend them outward alternately, repeat motion", "segment_frame_end": 120, "segment_frame_start": 0, "segment_index": 0, "segment_style": "fluid", "sequence_name": "0031_rom_poses_retarget__seg0001__part0001", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/g1/BioMotionLab_NTroje/rub055/0031_rom_poses_retarget__seg0001__part0001.npz", "split": "test", "text_path": null, "video_summary": "cross arms in front of body, then extend them outward alternately, repeat motion; style: fluid", "window_end": 90, "window_start": 30}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00056.npz", "sample_index": 56}
{"caption": "tilt torso and head side to side, bending upper body left and right alternately; style: smooth", "dataset": "amass_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 1936, "index": 1936, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00057.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/labels/BioMotionLab_NTroje/rub055/0031_rom_poses_retarget__seg0003__part0000.json", "segment_action": "tilt torso and head side to side, bending upper body left and right alternately", "segment_frame_end": 270, "segment_frame_start": 0, "segment_index": 0, "segment_style": "smooth", "sequence_name": "0031_rom_poses_retarget__seg0003__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/g1/BioMotionLab_NTroje/rub055/0031_rom_poses_retarget__seg0003__part0000.npz", "split": "test", "text_path": null, "video_summary": "tilt torso and head side to side, bending upper body left and right alternately; style: smooth", "window_end": 270, "window_start": 210}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00057.npz", "sample_index": 57}
{"caption": "stand still with arms extended horizontally in T-pose; style: static", "dataset": "amass_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 2172, "index": 2172, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00058.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/labels/BioMotionLab_NTroje/rub068/0031_rom_poses_retarget__seg0000__part0000.json", "segment_action": "stand still with arms extended horizontally in T-pose", "segment_frame_end": 210, "segment_frame_start": 0, "segment_index": 0, "segment_style": "static", "sequence_name": "0031_rom_poses_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/g1/BioMotionLab_NTroje/rub068/0031_rom_poses_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "stand still with arms extended horizontally in T-pose; style: static", "window_end": 135, "window_start": 75}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00058.npz", "sample_index": 58}
{"caption": "rotate 180 degrees, then walk away from the camera with arm swings; style: controlled", "dataset": "amass_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 2208, "index": 2208, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00059.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/labels/BioMotionLab_NTroje/rub069/0014_knocking2_poses_retarget__seg0002__part0000.json", "segment_action": "rotate 180 degrees, then walk away from the camera with arm swings", "segment_frame_end": 101, "segment_frame_start": 0, "segment_index": 0, "segment_style": "controlled", "sequence_name": "0014_knocking2_poses_retarget__seg0002__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/g1/BioMotionLab_NTroje/rub069/0014_knocking2_poses_retarget__seg0002__part0000.npz", "split": "test", "text_path": null, "video_summary": "rotate 180 degrees, then walk away from the camera with arm swings; style: controlled", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00059.npz", "sample_index": 59}
{"caption": "rotate torso, extend left arm outward, lift right leg high to execute a side kick while spinning; style: dynamic", "dataset": "amass_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 2284, "index": 2284, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00060.npz", "meta": {"eval_num_windows": 1, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/labels/BioMotionLab_NTroje/rub073/0026_kicking2_poses_retarget__seg0001__part0000.json", "segment_action": "rotate torso, extend left arm outward, lift right leg high to execute a side kick while spinning", "segment_frame_end": 60, "segment_frame_start": 0, "segment_index": 0, "segment_style": "dynamic", "sequence_name": "0026_kicking2_poses_retarget__seg0001__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/g1/BioMotionLab_NTroje/rub073/0026_kicking2_poses_retarget__seg0001__part0000.npz", "split": "test", "text_path": null, "video_summary": "rotate torso, extend left arm outward, lift right leg high to execute a side kick while spinning; style: dynamic", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00060.npz", "sample_index": 60}
{"caption": "walk forward with alternating leg steps, swing arms naturally, maintain steady gait; style: controlled, rhythmic", "dataset": "amass_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 2850, "index": 2850, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00061.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/labels/BioMotionLab_NTroje/rub111/0002_treadmill_slow_poses_retarget__seg0000__part0000.json", "segment_action": "walk forward with alternating leg steps, swing arms naturally, maintain steady gait", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "controlled, rhythmic", "sequence_name": "0002_treadmill_slow_poses_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/g1/BioMotionLab_NTroje/rub111/0002_treadmill_slow_poses_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "walk forward with alternating leg steps, swing arms naturally, maintain steady gait; style: controlled, rhythmic", "window_end": 180, "window_start": 120}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00061.npz", "sample_index": 61}
{"caption": "stand in place with arms extended horizontally to both sides, hold pose briefly; style: static", "dataset": "amass_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 2993, "index": 2993, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00062.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/labels/DanceDB/20120807_VasoAristeidou/Vasso_Bachata_01_poses_retarget__seg0004__part0000.json", "segment_action": "stand in place with arms extended horizontally to both sides, hold pose briefly", "segment_frame_end": 90, "segment_frame_start": 0, "segment_index": 0, "segment_style": "static", "sequence_name": "Vasso_Bachata_01_poses_retarget__seg0004__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/g1/DanceDB/20120807_VasoAristeidou/Vasso_Bachata_01_poses_retarget__seg0004__part0000.npz", "split": "test", "text_path": null, "video_summary": "stand in place with arms extended horizontally to both sides, hold pose briefly; style: static", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00062.npz", "sample_index": 62}
{"caption": "perform coordinated footwork, arm gestures, and body twists; style: rhythmic, energetic", "dataset": "amass_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 3150, "index": 3150, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00063.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/labels/DanceDB/20140526_StephanosKoullapis/StefanosKoullapis_Reggaeton_C3D_poses_retarget__seg0004__part0002.json", "segment_action": "perform coordinated footwork, arm gestures, and body twists", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "rhythmic, energetic", "sequence_name": "StefanosKoullapis_Reggaeton_C3D_poses_retarget__seg0004__part0002", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/g1/DanceDB/20140526_StephanosKoullapis/StefanosKoullapis_Reggaeton_C3D_poses_retarget__seg0004__part0002.npz", "split": "test", "text_path": null, "video_summary": "perform coordinated footwork, arm gestures, and body twists; style: rhythmic, energetic", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00063.npz", "sample_index": 63}
{"caption": "spin around rapidly while alternately lifting legs and swinging arms for balance; style: dynamic", "dataset": "amass_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 3397, "index": 3397, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00064.npz", "meta": {"eval_num_windows": 1, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/labels/EKUT/234/SLPD206_poses_retarget__seg0001__part0000.json", "segment_action": "spin around rapidly while alternately lifting legs and swinging arms for balance", "segment_frame_end": 60, "segment_frame_start": 0, "segment_index": 0, "segment_style": "dynamic", "sequence_name": "SLPD206_poses_retarget__seg0001__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/g1/EKUT/234/SLPD206_poses_retarget__seg0001__part0000.npz", "split": "test", "text_path": null, "video_summary": "spin around rapidly while alternately lifting legs and swinging arms for balance; style: dynamic", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00064.npz", "sample_index": 64}
{"caption": "lower arms, bend knees, rotate torso, raise arms overhead, then place hands on hips while shifting body weight; style: transitional", "dataset": "amass_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 3496, "index": 3496, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00065.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/labels/Eyes_Japan_Dataset/aita/accident-03-damagehead-aita_poses_retarget__seg0001__part0000.json", "segment_action": "lower arms, bend knees, rotate torso, raise arms overhead, then place hands on hips while shifting body weight", "segment_frame_end": 90, "segment_frame_start": 0, "segment_index": 0, "segment_style": "transitional", "sequence_name": "accident-03-damagehead-aita_poses_retarget__seg0001__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/g1/Eyes_Japan_Dataset/aita/accident-03-damagehead-aita_poses_retarget__seg0001__part0000.npz", "split": "test", "text_path": null, "video_summary": "lower arms, bend knees, rotate torso, raise arms overhead, then place hands on hips while shifting body weight; style: transitional", "window_end": 75, "window_start": 15}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00065.npz", "sample_index": 65}
{"caption": "cross right arm over left, sway hips side to side, and step in place with knee bends; style: rhythmic", "dataset": "amass_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 3684, "index": 3684, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00066.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/labels/Eyes_Japan_Dataset/hamada/accident-07-damageleftarm-hamada_poses_retarget__seg0002__part0000.json", "segment_action": "cross right arm over left, sway hips side to side, and step in place with knee bends", "segment_frame_end": 150, "segment_frame_start": 0, "segment_index": 0, "segment_style": "rhythmic", "sequence_name": "accident-07-damageleftarm-hamada_poses_retarget__seg0002__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/g1/Eyes_Japan_Dataset/hamada/accident-07-damageleftarm-hamada_poses_retarget__seg0002__part0000.npz", "split": "test", "text_path": null, "video_summary": "cross right arm over left, sway hips side to side, and step in place with knee bends; style: rhythmic", "window_end": 150, "window_start": 90}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00066.npz", "sample_index": 66}
{"caption": "take small forward steps, lower arms to sides, and shift body weight between legs; style: controlled", "dataset": "amass_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 3722, "index": 3722, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00067.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/labels/Eyes_Japan_Dataset/hamada/gesture_etc-25-magic-hamada_poses_retarget__seg0004__part0000.json", "segment_action": "take small forward steps, lower arms to sides, and shift body weight between legs", "segment_frame_end": 105, "segment_frame_start": 0, "segment_index": 0, "segment_style": "controlled", "sequence_name": "gesture_etc-25-magic-hamada_poses_retarget__seg0004__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/g1/Eyes_Japan_Dataset/hamada/gesture_etc-25-magic-hamada_poses_retarget__seg0004__part0000.npz", "split": "test", "text_path": null, "video_summary": "take small forward steps, lower arms to sides, and shift body weight between legs; style: controlled", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00067.npz", "sample_index": 67}
{"caption": "raise arms back to extended horizontal position and stand still; style: static", "dataset": "amass_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 4078, "index": 4078, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00068.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/labels/Eyes_Japan_Dataset/kaiwa/gesture_etc-46-vomitofblood-kaiwa_poses_retarget__seg0005__part0000.json", "segment_action": "raise arms back to extended horizontal position and stand still", "segment_frame_end": 90, "segment_frame_start": 0, "segment_index": 0, "segment_style": "static", "sequence_name": "gesture_etc-46-vomitofblood-kaiwa_poses_retarget__seg0005__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/g1/Eyes_Japan_Dataset/kaiwa/gesture_etc-46-vomitofblood-kaiwa_poses_retarget__seg0005__part0000.npz", "split": "test", "text_path": null, "video_summary": "raise arms back to extended horizontal position and stand still; style: static", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00068.npz", "sample_index": 68}
{"caption": "perform fine finger movements and hand manipulations while hands are held in front of torso; style: precise", "dataset": "amass_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 5120, "index": 5120, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00069.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/labels/GRAB/s4/hand_inspect_1_stageii_retarget_poses_retarget__seg0001__part0000.json", "segment_action": "perform fine finger movements and hand manipulations while hands are held in front of torso", "segment_frame_end": 180, "segment_frame_start": 0, "segment_index": 0, "segment_style": "precise", "sequence_name": "hand_inspect_1_stageii_retarget_poses_retarget__seg0001__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/g1/GRAB/s4/hand_inspect_1_stageii_retarget_poses_retarget__seg0001__part0000.npz", "split": "test", "text_path": null, "video_summary": "perform fine finger movements and hand manipulations while hands are held in front of torso; style: precise", "window_end": 180, "window_start": 120}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00069.npz", "sample_index": 69}
{"caption": "raise left arm to outward position, rotate torso left, then lower left arm back to the side; style: controlled", "dataset": "amass_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 5151, "index": 5151, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00070.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/labels/GRAB/s4/watch_pass_1_stageii_retarget_poses_retarget__seg0001__part0000.json", "segment_action": "raise left arm to outward position, rotate torso left, then lower left arm back to the side", "segment_frame_end": 66, "segment_frame_start": 0, "segment_index": 0, "segment_style": "controlled", "sequence_name": "watch_pass_1_stageii_retarget_poses_retarget__seg0001__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/g1/GRAB/s4/watch_pass_1_stageii_retarget_poses_retarget__seg0001__part0000.npz", "split": "test", "text_path": null, "video_summary": "raise left arm to outward position, rotate torso left, then lower left arm back to the side; style: controlled", "window_end": 66, "window_start": 6}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00070.npz", "sample_index": 70}
{"caption": "stand in place, extend left arm outward then retract it, keep right hand on hip; style: steady", "dataset": "amass_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 5232, "index": 5232, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00071.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/labels/GRAB/s5/stanfordbunny_pick_all_stageii_retarget_poses_retarget__seg0003__part0000.json", "segment_action": "stand in place, extend left arm outward then retract it, keep right hand on hip", "segment_frame_end": 210, "segment_frame_start": 0, "segment_index": 0, "segment_style": "steady", "sequence_name": "stanfordbunny_pick_all_stageii_retarget_poses_retarget__seg0003__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/g1/GRAB/s5/stanfordbunny_pick_all_stageii_retarget_poses_retarget__seg0003__part0000.npz", "split": "test", "text_path": null, "video_summary": "stand in place, extend left arm outward then retract it, keep right hand on hip; style: steady", "window_end": 135, "window_start": 75}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00071.npz", "sample_index": 71}
{"caption": "extend left arm horizontally outward, hold position briefly, then retract arm back to side; style: deliberate", "dataset": "amass_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 5624, "index": 5624, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00072.npz", "meta": {"eval_num_windows": 1, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/labels/GRAB/s9/mouse_pass_1_stageii_retarget_poses_retarget__seg0001__part0000.json", "segment_action": "extend left arm horizontally outward, hold position briefly, then retract arm back to side", "segment_frame_end": 60, "segment_frame_start": 0, "segment_index": 0, "segment_style": "deliberate", "sequence_name": "mouse_pass_1_stageii_retarget_poses_retarget__seg0001__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/g1/GRAB/s9/mouse_pass_1_stageii_retarget_poses_retarget__seg0001__part0000.npz", "split": "test", "text_path": null, "video_summary": "extend left arm horizontally outward, hold position briefly, then retract arm back to side; style: deliberate", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00072.npz", "sample_index": 72}
{"caption": "rotate torso to the left, extend both arms forward in a reaching and grabbing motion; style: controlled", "dataset": "amass_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 5645, "index": 5645, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00073.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/labels/GRAB/s9/waterbottle_pick_all_stageii_retarget_poses_retarget__seg0002__part0000.json", "segment_action": "rotate torso to the left, extend both arms forward in a reaching and grabbing motion", "segment_frame_end": 150, "segment_frame_start": 0, "segment_index": 0, "segment_style": "controlled", "sequence_name": "waterbottle_pick_all_stageii_retarget_poses_retarget__seg0002__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/g1/GRAB/s9/waterbottle_pick_all_stageii_retarget_poses_retarget__seg0002__part0000.npz", "split": "test", "text_path": null, "video_summary": "rotate torso to the left, extend both arms forward in a reaching and grabbing motion; style: controlled", "window_end": 105, "window_start": 45}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00073.npz", "sample_index": 73}
{"caption": "sit with legs bent, lean back, raise arms, lift legs slightly and shift body side to side; style: dynamic", "dataset": "amass_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 5759, "index": 5759, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00074.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/labels/HUMAN4D/Subject3_4_Carine/RGB_WatchingFootball_S3S4_01_poses_retarget__seg0006__part0000.json", "segment_action": "sit with legs bent, lean back, raise arms, lift legs slightly and shift body side to side", "segment_frame_end": 155, "segment_frame_start": 0, "segment_index": 0, "segment_style": "dynamic", "sequence_name": "RGB_WatchingFootball_S3S4_01_poses_retarget__seg0006__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/g1/HUMAN4D/Subject3_4_Carine/RGB_WatchingFootball_S3S4_01_poses_retarget__seg0006__part0000.npz", "split": "test", "text_path": null, "video_summary": "sit with legs bent, lean back, raise arms, lift legs slightly and shift body side to side; style: dynamic", "window_end": 108, "window_start": 48}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00074.npz", "sample_index": 74}
{"caption": "hold left arm up briefly then lower it slowly back to hip position; style: smooth", "dataset": "amass_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 6043, "index": 6043, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00075.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/labels/KIT/1229/take_of_t-shirt_with_help_03_poses_retarget__seg0002__part0000.json", "segment_action": "hold left arm up briefly then lower it slowly back to hip position", "segment_frame_end": 63, "segment_frame_start": 0, "segment_index": 0, "segment_style": "smooth", "sequence_name": "take_of_t-shirt_with_help_03_poses_retarget__seg0002__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/g1/KIT/1229/take_of_t-shirt_with_help_03_poses_retarget__seg0002__part0000.npz", "split": "test", "text_path": null, "video_summary": "hold left arm up briefly then lower it slowly back to hip position; style: smooth", "window_end": 62, "window_start": 2}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00075.npz", "sample_index": 75}
{"caption": "stand with arms extended horizontally to both sides, perform repeated small vertical jumps; style: controlled", "dataset": "amass_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 6065, "index": 6065, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00076.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/labels/KIT/1346/Experiment3_subject1346_wash_back_position_horizontal_01_poses_retarget__seg0000__part0000.json", "segment_action": "stand with arms extended horizontally to both sides, perform repeated small vertical jumps", "segment_frame_end": 150, "segment_frame_start": 0, "segment_index": 0, "segment_style": "controlled", "sequence_name": "Experiment3_subject1346_wash_back_position_horizontal_01_poses_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/g1/KIT/1346/Experiment3_subject1346_wash_back_position_horizontal_01_poses_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "stand with arms extended horizontally to both sides, perform repeated small vertical jumps; style: controlled", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00076.npz", "sample_index": 76}
{"caption": "shift body weight, extend arms outward, and perform a lateral displacement step to the right; style: smooth", "dataset": "amass_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 6343, "index": 6343, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00077.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/labels/KIT/1747/displace_from_left_to_right_44_poses_retarget__seg0001__part0000.json", "segment_action": "shift body weight, extend arms outward, and perform a lateral displacement step to the right", "segment_frame_end": 105, "segment_frame_start": 0, "segment_index": 0, "segment_style": "smooth", "sequence_name": "displace_from_left_to_right_44_poses_retarget__seg0001__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/g1/KIT/1747/displace_from_left_to_right_44_poses_retarget__seg0001__part0000.npz", "split": "test", "text_path": null, "video_summary": "shift body weight, extend arms outward, and perform a lateral displacement step to the right; style: smooth", "window_end": 105, "window_start": 45}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00077.npz", "sample_index": 77}
{"caption": "step in a circular path while rotating 180 degrees to the right, arms swing naturally with body movement; style: controlled", "dataset": "amass_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 6411, "index": 6411, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00078.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/labels/KIT/183/turn_right04_poses_retarget__seg0001__part0000.json", "segment_action": "step in a circular path while rotating 180 degrees to the right, arms swing naturally with body movement", "segment_frame_end": 105, "segment_frame_start": 0, "segment_index": 0, "segment_style": "controlled", "sequence_name": "turn_right04_poses_retarget__seg0001__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/g1/KIT/183/turn_right04_poses_retarget__seg0001__part0000.npz", "split": "test", "text_path": null, "video_summary": "step in a circular path while rotating 180 degrees to the right, arms swing naturally with body movement; style: controlled", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00078.npz", "sample_index": 78}
{"caption": "walk forward with coordinated arm and leg movements; style: smooth", "dataset": "amass_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 6859, "index": 6859, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00079.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/labels/KIT/359/bend_left10_poses_retarget__seg0001__part0000.json", "segment_action": "walk forward with coordinated arm and leg movements", "segment_frame_end": 81, "segment_frame_start": 0, "segment_index": 0, "segment_style": "smooth", "sequence_name": "bend_left10_poses_retarget__seg0001__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/g1/KIT/359/bend_left10_poses_retarget__seg0001__part0000.npz", "split": "test", "text_path": null, "video_summary": "walk forward with coordinated arm and leg movements; style: smooth", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00079.npz", "sample_index": 79}
{"caption": "turn toward camera, step forward repeatedly to recover stability; style: dynamic", "dataset": "amass_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 6954, "index": 6954, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00080.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/labels/KIT/421/push_recovery_right03_poses_retarget__seg0002__part0000.json", "segment_action": "turn toward camera, step forward repeatedly to recover stability", "segment_frame_end": 65, "segment_frame_start": 0, "segment_index": 0, "segment_style": "dynamic", "sequence_name": "push_recovery_right03_poses_retarget__seg0002__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/g1/KIT/421/push_recovery_right03_poses_retarget__seg0002__part0000.npz", "split": "test", "text_path": null, "video_summary": "turn toward camera, step forward repeatedly to recover stability; style: dynamic", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00080.npz", "sample_index": 80}
{"caption": "shift body weight between legs, raise right arm upward and lower back down, rotate torso slightly to the left; style: controlled, slow", "dataset": "amass_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 7131, "index": 7131, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00081.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/labels/KIT/442/PizzaDelivery04_poses_retarget__seg0001__part0000.json", "segment_action": "shift body weight between legs, raise right arm upward and lower back down, rotate torso slightly to the left", "segment_frame_end": 153, "segment_frame_start": 0, "segment_index": 0, "segment_style": "controlled, slow", "sequence_name": "PizzaDelivery04_poses_retarget__seg0001__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/g1/KIT/442/PizzaDelivery04_poses_retarget__seg0001__part0000.npz", "split": "test", "text_path": null, "video_summary": "shift body weight between legs, raise right arm upward and lower back down, rotate torso slightly to the left; style: controlled, slow", "window_end": 106, "window_start": 46}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00081.npz", "sample_index": 81}
{"caption": "walk slowly while maintaining balance, swing arms to stabilize the body; style: controlled", "dataset": "amass_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 7170, "index": 7170, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00082.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/labels/KIT/513/balance_on_beam01_poses_retarget__seg0001__part0000.json", "segment_action": "walk slowly while maintaining balance, swing arms to stabilize the body", "segment_frame_end": 143, "segment_frame_start": 0, "segment_index": 0, "segment_style": "controlled", "sequence_name": "balance_on_beam01_poses_retarget__seg0001__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/g1/KIT/513/balance_on_beam01_poses_retarget__seg0001__part0000.npz", "split": "test", "text_path": null, "video_summary": "walk slowly while maintaining balance, swing arms to stabilize the body; style: controlled", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00082.npz", "sample_index": 82}
{"caption": "rotate torso left, step forward with right leg, swing arms to maintain balance; style: controlled", "dataset": "amass_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 7226, "index": 7226, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00083.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/labels/KIT/551/sponge_big07_poses_retarget__seg0000__part0000.json", "segment_action": "rotate torso left, step forward with right leg, swing arms to maintain balance", "segment_frame_end": 66, "segment_frame_start": 0, "segment_index": 0, "segment_style": "controlled", "sequence_name": "sponge_big07_poses_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/g1/KIT/551/sponge_big07_poses_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "rotate torso left, step forward with right leg, swing arms to maintain balance; style: controlled", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00083.npz", "sample_index": 83}
{"caption": "lower extended arms to sides, take two forward steps, and stand still; style: controlled", "dataset": "amass_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 7557, "index": 7557, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00084.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/labels/KIT/675/Trial_02_poses_retarget__seg0000__part0000.json", "segment_action": "lower extended arms to sides, take two forward steps, and stand still", "segment_frame_end": 90, "segment_frame_start": 0, "segment_index": 0, "segment_style": "controlled", "sequence_name": "Trial_02_poses_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/g1/KIT/675/Trial_02_poses_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "lower extended arms to sides, take two forward steps, and stand still; style: controlled", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00084.npz", "sample_index": 84}
{"caption": "rotate body 180 degrees while stepping with legs, lower arms from hips to sides; style: controlled", "dataset": "amass_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 7807, "index": 7807, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00085.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/labels/KIT/9/bend_right07_poses_retarget__seg0000__part0000.json", "segment_action": "rotate body 180 degrees while stepping with legs, lower arms from hips to sides", "segment_frame_end": 75, "segment_frame_start": 0, "segment_index": 0, "segment_style": "controlled", "sequence_name": "bend_right07_poses_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/g1/KIT/9/bend_right07_poses_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "rotate body 180 degrees while stepping with legs, lower arms from hips to sides; style: controlled", "window_end": 68, "window_start": 8}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00085.npz", "sample_index": 85}
{"caption": "raise right arm, perform circular motion above head, then lower arm to side; style: controlled", "dataset": "amass_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 7939, "index": 7939, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00086.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/labels/KIT/969/Trial_06_poses_retarget__seg0002__part0000.json", "segment_action": "raise right arm, perform circular motion above head, then lower arm to side", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "controlled", "sequence_name": "Trial_06_poses_retarget__seg0002__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/g1/KIT/969/Trial_06_poses_retarget__seg0002__part0000.npz", "split": "test", "text_path": null, "video_summary": "raise right arm, perform circular motion above head, then lower arm to side; style: controlled", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00086.npz", "sample_index": 86}
{"caption": "extend both arms forward, then retract them back to rest on legs; style: dynamic", "dataset": "amass_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 8074, "index": 8074, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00087.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/labels/KIT/987/Experiment3a_07_poses_retarget__seg0004__part0000.json", "segment_action": "extend both arms forward, then retract them back to rest on legs", "segment_frame_end": 90, "segment_frame_start": 0, "segment_index": 0, "segment_style": "dynamic", "sequence_name": "Experiment3a_07_poses_retarget__seg0004__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/g1/KIT/987/Experiment3a_07_poses_retarget__seg0004__part0000.npz", "split": "test", "text_path": null, "video_summary": "extend both arms forward, then retract them back to rest on legs; style: dynamic", "window_end": 75, "window_start": 15}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00087.npz", "sample_index": 87}
{"caption": "extend both arms forward, then retract back to legs; style: dynamic", "dataset": "amass_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 8093, "index": 8093, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00088.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/labels/KIT/987/Experiment3a_07_poses_retarget__seg0010__part0000.json", "segment_action": "extend both arms forward, then retract back to legs", "segment_frame_end": 120, "segment_frame_start": 0, "segment_index": 0, "segment_style": "dynamic", "sequence_name": "Experiment3a_07_poses_retarget__seg0010__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/g1/KIT/987/Experiment3a_07_poses_retarget__seg0010__part0000.npz", "split": "test", "text_path": null, "video_summary": "extend both arms forward, then retract back to legs; style: dynamic", "window_end": 120, "window_start": 60}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00088.npz", "sample_index": 88}
{"caption": "step in place with quick foot movements, rotating body and moving arms in rhythmic pattern; style: rhythmic", "dataset": "amass_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 8137, "index": 8137, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00089.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/labels/MPI_HDM05/bk/HDM_bk_01-01_01_120_poses_retarget__seg0005__part0000.json", "segment_action": "step in place with quick foot movements, rotating body and moving arms in rhythmic pattern", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "rhythmic", "sequence_name": "HDM_bk_01-01_01_120_poses_retarget__seg0005__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/g1/MPI_HDM05/bk/HDM_bk_01-01_01_120_poses_retarget__seg0005__part0000.npz", "split": "test", "text_path": null, "video_summary": "step in place with quick foot movements, rotating body and moving arms in rhythmic pattern; style: rhythmic", "window_end": 180, "window_start": 120}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00089.npz", "sample_index": 89}
{"caption": "execute alternating high side kicks, using arms for balance; style: dynamic", "dataset": "amass_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 8163, "index": 8163, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00090.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/labels/MPI_HDM05/bk/HDM_bk_03-05_01_120_poses_retarget__seg0006__part0000.json", "segment_action": "execute alternating high side kicks, using arms for balance", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "dynamic", "sequence_name": "HDM_bk_03-05_01_120_poses_retarget__seg0006__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/g1/MPI_HDM05/bk/HDM_bk_03-05_01_120_poses_retarget__seg0006__part0000.npz", "split": "test", "text_path": null, "video_summary": "execute alternating high side kicks, using arms for balance; style: dynamic", "window_end": 300, "window_start": 240}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00090.npz", "sample_index": 90}
{"caption": "spin clockwise on one foot, raise right arm overhead, extend left arm outward, and keep legs bent during rotation; style: smooth, rhythmic", "dataset": "amass_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 8488, "index": 8488, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00091.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/labels/MPI_mosh/50007/misc1_poses_retarget__seg0000__part0000.json", "segment_action": "spin clockwise on one foot, raise right arm overhead, extend left arm outward, and keep legs bent during rotation", "segment_frame_end": 93, "segment_frame_start": 0, "segment_index": 0, "segment_style": "smooth, rhythmic", "sequence_name": "misc1_poses_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/g1/MPI_mosh/50007/misc1_poses_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "spin clockwise on one foot, raise right arm overhead, extend left arm outward, and keep legs bent during rotation; style: smooth, rhythmic", "window_end": 76, "window_start": 16}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00091.npz", "sample_index": 91}
{"caption": "lean forward slightly, rotate torso sharply side to side, adjust leg position for balance; style: controlled", "dataset": "amass_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 8517, "index": 8517, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00092.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/labels/MPI_mosh/50021/hips_poses_retarget__seg0002__part0000.json", "segment_action": "lean forward slightly, rotate torso sharply side to side, adjust leg position for balance", "segment_frame_end": 90, "segment_frame_start": 0, "segment_index": 0, "segment_style": "controlled", "sequence_name": "hips_poses_retarget__seg0002__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/g1/MPI_mosh/50021/hips_poses_retarget__seg0002__part0000.npz", "split": "test", "text_path": null, "video_summary": "lean forward slightly, rotate torso sharply side to side, adjust leg position for balance; style: controlled", "window_end": 75, "window_start": 15}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00092.npz", "sample_index": 92}
{"caption": "rise from crouched position to standing, straighten legs, and extend arms outward to return to initial stance; style: smooth", "dataset": "amass_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 8932, "index": 8932, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00093.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/labels/Transitions_mocap/mazen_c3d/crawl_stand_poses_retarget__seg0002__part0000.json", "segment_action": "rise from crouched position to standing, straighten legs, and extend arms outward to return to initial stance", "segment_frame_end": 75, "segment_frame_start": 0, "segment_index": 0, "segment_style": "smooth", "sequence_name": "crawl_stand_poses_retarget__seg0002__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/amass/g1/Transitions_mocap/mazen_c3d/crawl_stand_poses_retarget__seg0002__part0000.npz", "split": "test", "text_path": null, "video_summary": "rise from crouched position to standing, straighten legs, and extend arms outward to return to initial stance; style: smooth", "window_end": 68, "window_start": 8}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00093.npz", "sample_index": 93}
{"caption": "", "dataset": "beat2_original_humanref_test", "fps": 30.0, "frames": 60, "global_index": 164968, "index": 1, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00094.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": null, "segment_action": "giving a speech", "segment_frame_end": 2423, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "15_carlos_4_66_66_retarget", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/beat2_spanish/g1/15_carlos_4_66_66_retarget.npz", "split": "test", "text_path": null, "video_summary": "giving a speech", "window_end": 1242, "window_start": 1182}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00094.npz", "sample_index": 94}
{"caption": "", "dataset": "beat2_original_humanref_test", "fps": 30.0, "frames": 60, "global_index": 164970, "index": 3, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00095.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": null, "segment_action": "giving a speech", "segment_frame_end": 2213, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "15_carlos_4_74_74_retarget", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/beat2_spanish/g1/15_carlos_4_74_74_retarget.npz", "split": "test", "text_path": null, "video_summary": "giving a speech", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00095.npz", "sample_index": 95}
{"caption": "", "dataset": "beat2_original_humanref_test", "fps": 30.0, "frames": 60, "global_index": 164971, "index": 4, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00096.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": null, "segment_action": "giving a speech", "segment_frame_end": 2213, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "15_carlos_4_74_74_retarget", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/beat2_spanish/g1/15_carlos_4_74_74_retarget.npz", "split": "test", "text_path": null, "video_summary": "giving a speech", "window_end": 1136, "window_start": 1076}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00096.npz", "sample_index": 96}
{"caption": "", "dataset": "beat2_original_humanref_test", "fps": 30.0, "frames": 60, "global_index": 164973, "index": 6, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00097.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": null, "segment_action": "giving a speech", "segment_frame_end": 1490, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "16_jorge_4_111_111_retarget", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/beat2_spanish/g1/16_jorge_4_111_111_retarget.npz", "split": "test", "text_path": null, "video_summary": "giving a speech", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00097.npz", "sample_index": 97}
{"caption": "", "dataset": "beat2_original_humanref_test", "fps": 30.0, "frames": 60, "global_index": 164976, "index": 9, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00098.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": null, "segment_action": "giving a speech", "segment_frame_end": 2140, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "17_itoi_6_112_112_retarget", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/beat2_japanese/g1/17_itoi_6_112_112_retarget.npz", "split": "test", "text_path": null, "video_summary": "giving a speech", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00098.npz", "sample_index": 98}
{"caption": "", "dataset": "beat2_original_humanref_test", "fps": 30.0, "frames": 60, "global_index": 164978, "index": 11, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00099.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": null, "segment_action": "giving a speech", "segment_frame_end": 2140, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "17_itoi_6_112_112_retarget", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/beat2_japanese/g1/17_itoi_6_112_112_retarget.npz", "split": "test", "text_path": null, "video_summary": "giving a speech", "window_end": 2140, "window_start": 2080}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00099.npz", "sample_index": 99}
{"caption": "", "dataset": "beat2_original_humanref_test", "fps": 30.0, "frames": 60, "global_index": 164979, "index": 12, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00100.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": null, "segment_action": "giving a speech", "segment_frame_end": 2303, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "17_itoi_6_14_14_retarget", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/beat2_japanese/g1/17_itoi_6_14_14_retarget.npz", "split": "test", "text_path": null, "video_summary": "giving a speech", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00100.npz", "sample_index": 100}
{"caption": "", "dataset": "beat2_original_humanref_test", "fps": 30.0, "frames": 60, "global_index": 164980, "index": 13, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00101.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": null, "segment_action": "giving a speech", "segment_frame_end": 2303, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "17_itoi_6_14_14_retarget", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/beat2_japanese/g1/17_itoi_6_14_14_retarget.npz", "split": "test", "text_path": null, "video_summary": "giving a speech", "window_end": 1182, "window_start": 1122}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00101.npz", "sample_index": 101}
{"caption": "", "dataset": "beat2_original_humanref_test", "fps": 30.0, "frames": 60, "global_index": 164983, "index": 16, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00102.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": null, "segment_action": "giving a speech", "segment_frame_end": 2033, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "17_itoi_6_7_7_retarget", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/beat2_japanese/g1/17_itoi_6_7_7_retarget.npz", "split": "test", "text_path": null, "video_summary": "giving a speech", "window_end": 1046, "window_start": 986}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00102.npz", "sample_index": 102}
{"caption": "", "dataset": "beat2_original_humanref_test", "fps": 30.0, "frames": 60, "global_index": 164985, "index": 18, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00103.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": null, "segment_action": "giving a speech", "segment_frame_end": 2339, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "25_goto_6_14_14_retarget", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/beat2_japanese/g1/25_goto_6_14_14_retarget.npz", "split": "test", "text_path": null, "video_summary": "giving a speech", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00103.npz", "sample_index": 103}
{"caption": "", "dataset": "beat2_original_humanref_test", "fps": 30.0, "frames": 60, "global_index": 164987, "index": 20, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00104.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": null, "segment_action": "giving a speech", "segment_frame_end": 2339, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "25_goto_6_14_14_retarget", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/beat2_japanese/g1/25_goto_6_14_14_retarget.npz", "split": "test", "text_path": null, "video_summary": "giving a speech", "window_end": 2339, "window_start": 2279}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00104.npz", "sample_index": 104}
{"caption": "", "dataset": "beat2_original_humanref_test", "fps": 30.0, "frames": 60, "global_index": 164988, "index": 21, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00105.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": null, "segment_action": "giving a speech", "segment_frame_end": 2347, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "25_goto_6_74_74_retarget", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/beat2_japanese/g1/25_goto_6_74_74_retarget.npz", "split": "test", "text_path": null, "video_summary": "giving a speech", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00105.npz", "sample_index": 105}
{"caption": "", "dataset": "beat2_original_humanref_test", "fps": 30.0, "frames": 60, "global_index": 164989, "index": 22, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00106.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": null, "segment_action": "giving a speech", "segment_frame_end": 2347, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "25_goto_6_74_74_retarget", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/beat2_japanese/g1/25_goto_6_74_74_retarget.npz", "split": "test", "text_path": null, "video_summary": "giving a speech", "window_end": 1204, "window_start": 1144}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00106.npz", "sample_index": 106}
{"caption": "", "dataset": "beat2_original_humanref_test", "fps": 30.0, "frames": 60, "global_index": 164994, "index": 27, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00107.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": null, "segment_action": "giving a speech", "segment_frame_end": 2063, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "10_kieks_0_2_2_retarget", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/beat2_english/g1/10_kieks_0_2_2_retarget.npz", "split": "test", "text_path": null, "video_summary": "giving a speech", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00107.npz", "sample_index": 107}
{"caption": "", "dataset": "beat2_original_humanref_test", "fps": 30.0, "frames": 60, "global_index": 164997, "index": 30, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00108.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": null, "segment_action": "giving a speech", "segment_frame_end": 1613, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "10_kieks_0_95_95_retarget", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/beat2_english/g1/10_kieks_0_95_95_retarget.npz", "split": "test", "text_path": null, "video_summary": "giving a speech", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00108.npz", "sample_index": 108}
{"caption": "", "dataset": "beat2_original_humanref_test", "fps": 30.0, "frames": 60, "global_index": 164998, "index": 31, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00109.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": null, "segment_action": "giving a speech", "segment_frame_end": 1613, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "10_kieks_0_95_95_retarget", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/beat2_english/g1/10_kieks_0_95_95_retarget.npz", "split": "test", "text_path": null, "video_summary": "giving a speech", "window_end": 836, "window_start": 776}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00109.npz", "sample_index": 109}
{"caption": "", "dataset": "beat2_original_humanref_test", "fps": 30.0, "frames": 60, "global_index": 165000, "index": 33, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00110.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": null, "segment_action": "giving a speech", "segment_frame_end": 1703, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "10_kieks_0_9_9_retarget", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/beat2_english/g1/10_kieks_0_9_9_retarget.npz", "split": "test", "text_path": null, "video_summary": "giving a speech", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00110.npz", "sample_index": 110}
{"caption": "", "dataset": "beat2_original_humanref_test", "fps": 30.0, "frames": 60, "global_index": 165002, "index": 35, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00111.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": null, "segment_action": "giving a speech", "segment_frame_end": 1703, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "10_kieks_0_9_9_retarget", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/beat2_english/g1/10_kieks_0_9_9_retarget.npz", "split": "test", "text_path": null, "video_summary": "giving a speech", "window_end": 1703, "window_start": 1643}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00111.npz", "sample_index": 111}
{"caption": "", "dataset": "beat2_original_humanref_test", "fps": 30.0, "frames": 60, "global_index": 165004, "index": 37, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00112.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": null, "segment_action": "giving a speech", "segment_frame_end": 3174, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "11_nidal_0_12_12_retarget", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/beat2_english/g1/11_nidal_0_12_12_retarget.npz", "split": "test", "text_path": null, "video_summary": "giving a speech", "window_end": 1617, "window_start": 1557}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00112.npz", "sample_index": 112}
{"caption": "", "dataset": "beat2_original_humanref_test", "fps": 30.0, "frames": 60, "global_index": 165011, "index": 44, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00113.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": null, "segment_action": "giving a speech", "segment_frame_end": 2273, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "11_nidal_0_50_50_retarget", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/beat2_english/g1/11_nidal_0_50_50_retarget.npz", "split": "test", "text_path": null, "video_summary": "giving a speech", "window_end": 2273, "window_start": 2213}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00113.npz", "sample_index": 113}
{"caption": "", "dataset": "beat2_original_humanref_test", "fps": 30.0, "frames": 60, "global_index": 165012, "index": 45, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00114.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": null, "segment_action": "giving a speech", "segment_frame_end": 2093, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "11_nidal_0_63_63_retarget", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/beat2_english/g1/11_nidal_0_63_63_retarget.npz", "split": "test", "text_path": null, "video_summary": "giving a speech", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00114.npz", "sample_index": 114}
{"caption": "", "dataset": "beat2_original_humanref_test", "fps": 30.0, "frames": 60, "global_index": 165014, "index": 47, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00115.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": null, "segment_action": "giving a speech", "segment_frame_end": 2093, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "11_nidal_0_63_63_retarget", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/beat2_english/g1/11_nidal_0_63_63_retarget.npz", "split": "test", "text_path": null, "video_summary": "giving a speech", "window_end": 2093, "window_start": 2033}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00115.npz", "sample_index": 115}
{"caption": "", "dataset": "beat2_original_humanref_test", "fps": 30.0, "frames": 60, "global_index": 165023, "index": 56, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00116.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": null, "segment_action": "giving a speech", "segment_frame_end": 2063, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "11_nidal_0_73_73_retarget", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/beat2_english/g1/11_nidal_0_73_73_retarget.npz", "split": "test", "text_path": null, "video_summary": "giving a speech", "window_end": 2063, "window_start": 2003}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00116.npz", "sample_index": 116}
{"caption": "", "dataset": "beat2_original_humanref_test", "fps": 30.0, "frames": 60, "global_index": 165025, "index": 58, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00117.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": null, "segment_action": "giving a speech", "segment_frame_end": 1823, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "11_nidal_0_76_76_retarget", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/beat2_english/g1/11_nidal_0_76_76_retarget.npz", "split": "test", "text_path": null, "video_summary": "giving a speech", "window_end": 942, "window_start": 882}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00117.npz", "sample_index": 117}
{"caption": "", "dataset": "beat2_original_humanref_test", "fps": 30.0, "frames": 60, "global_index": 165031, "index": 64, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00118.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": null, "segment_action": "giving a speech", "segment_frame_end": 2123, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "11_nidal_0_84_84_retarget", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/beat2_english/g1/11_nidal_0_84_84_retarget.npz", "split": "test", "text_path": null, "video_summary": "giving a speech", "window_end": 1092, "window_start": 1032}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00118.npz", "sample_index": 118}
{"caption": "", "dataset": "beat2_original_humanref_test", "fps": 30.0, "frames": 60, "global_index": 165036, "index": 69, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00119.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": null, "segment_action": "giving a speech", "segment_frame_end": 2094, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "11_nidal_0_9_9_retarget", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/beat2_english/g1/11_nidal_0_9_9_retarget.npz", "split": "test", "text_path": null, "video_summary": "giving a speech", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00119.npz", "sample_index": 119}
{"caption": "", "dataset": "beat2_original_humanref_test", "fps": 30.0, "frames": 60, "global_index": 165037, "index": 70, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00120.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": null, "segment_action": "giving a speech", "segment_frame_end": 2094, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "11_nidal_0_9_9_retarget", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/beat2_english/g1/11_nidal_0_9_9_retarget.npz", "split": "test", "text_path": null, "video_summary": "giving a speech", "window_end": 1077, "window_start": 1017}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00120.npz", "sample_index": 120}
{"caption": "", "dataset": "beat2_original_humanref_test", "fps": 30.0, "frames": 60, "global_index": 165043, "index": 76, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00121.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": null, "segment_action": "giving a speech", "segment_frame_end": 2273, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "12_zhao_0_4_4_retarget", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/beat2_english/g1/12_zhao_0_4_4_retarget.npz", "split": "test", "text_path": null, "video_summary": "giving a speech", "window_end": 1166, "window_start": 1106}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00121.npz", "sample_index": 121}
{"caption": "", "dataset": "beat2_original_humanref_test", "fps": 30.0, "frames": 60, "global_index": 165046, "index": 79, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00122.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": null, "segment_action": "giving a speech", "segment_frame_end": 2799, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "12_zhao_0_88_88_retarget", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/beat2_english/g1/12_zhao_0_88_88_retarget.npz", "split": "test", "text_path": null, "video_summary": "giving a speech", "window_end": 1430, "window_start": 1370}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00122.npz", "sample_index": 122}
{"caption": "", "dataset": "beat2_original_humanref_test", "fps": 30.0, "frames": 60, "global_index": 165047, "index": 80, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00123.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": null, "segment_action": "giving a speech", "segment_frame_end": 2799, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "12_zhao_0_88_88_retarget", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/beat2_english/g1/12_zhao_0_88_88_retarget.npz", "split": "test", "text_path": null, "video_summary": "giving a speech", "window_end": 2799, "window_start": 2739}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00123.npz", "sample_index": 123}
{"caption": "", "dataset": "beat2_original_humanref_test", "fps": 30.0, "frames": 60, "global_index": 165048, "index": 81, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00124.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": null, "segment_action": "giving a speech", "segment_frame_end": 1853, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "13_lu_0_10_10_retarget", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/beat2_english/g1/13_lu_0_10_10_retarget.npz", "split": "test", "text_path": null, "video_summary": "giving a speech", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00124.npz", "sample_index": 124}
{"caption": "", "dataset": "beat2_original_humanref_test", "fps": 30.0, "frames": 60, "global_index": 165049, "index": 82, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00125.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": null, "segment_action": "giving a speech", "segment_frame_end": 1853, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "13_lu_0_10_10_retarget", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/beat2_english/g1/13_lu_0_10_10_retarget.npz", "split": "test", "text_path": null, "video_summary": "giving a speech", "window_end": 956, "window_start": 896}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00125.npz", "sample_index": 125}
{"caption": "", "dataset": "beat2_original_humanref_test", "fps": 30.0, "frames": 60, "global_index": 165050, "index": 83, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00126.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": null, "segment_action": "giving a speech", "segment_frame_end": 1853, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "13_lu_0_10_10_retarget", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/beat2_english/g1/13_lu_0_10_10_retarget.npz", "split": "test", "text_path": null, "video_summary": "giving a speech", "window_end": 1853, "window_start": 1793}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00126.npz", "sample_index": 126}
{"caption": "", "dataset": "beat2_original_humanref_test", "fps": 30.0, "frames": 60, "global_index": 165051, "index": 84, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00127.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": null, "segment_action": "giving a speech", "segment_frame_end": 1463, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "13_lu_0_65_65_retarget", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/beat2_english/g1/13_lu_0_65_65_retarget.npz", "split": "test", "text_path": null, "video_summary": "giving a speech", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00127.npz", "sample_index": 127}
{"caption": "", "dataset": "beat2_original_humanref_test", "fps": 30.0, "frames": 60, "global_index": 165052, "index": 85, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00128.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": null, "segment_action": "giving a speech", "segment_frame_end": 1463, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "13_lu_0_65_65_retarget", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/beat2_english/g1/13_lu_0_65_65_retarget.npz", "split": "test", "text_path": null, "video_summary": "giving a speech", "window_end": 762, "window_start": 702}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00128.npz", "sample_index": 128}
{"caption": "", "dataset": "beat2_original_humanref_test", "fps": 30.0, "frames": 60, "global_index": 165053, "index": 86, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00129.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": null, "segment_action": "giving a speech", "segment_frame_end": 1463, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "13_lu_0_65_65_retarget", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/beat2_english/g1/13_lu_0_65_65_retarget.npz", "split": "test", "text_path": null, "video_summary": "giving a speech", "window_end": 1463, "window_start": 1403}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00129.npz", "sample_index": 129}
{"caption": "", "dataset": "beat2_original_humanref_test", "fps": 30.0, "frames": 60, "global_index": 165054, "index": 87, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00130.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": null, "segment_action": "giving a speech", "segment_frame_end": 2393, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "13_lu_0_82_82_retarget", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/beat2_english/g1/13_lu_0_82_82_retarget.npz", "split": "test", "text_path": null, "video_summary": "giving a speech", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00130.npz", "sample_index": 130}
{"caption": "", "dataset": "beat2_original_humanref_test", "fps": 30.0, "frames": 60, "global_index": 165058, "index": 91, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00131.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": null, "segment_action": "giving a speech", "segment_frame_end": 3083, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "15_carlos_0_15_15_retarget", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/beat2_english/g1/15_carlos_0_15_15_retarget.npz", "split": "test", "text_path": null, "video_summary": "giving a speech", "window_end": 1572, "window_start": 1512}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00131.npz", "sample_index": 131}
{"caption": "", "dataset": "beat2_original_humanref_test", "fps": 30.0, "frames": 60, "global_index": 165060, "index": 93, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00132.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": null, "segment_action": "giving a speech", "segment_frame_end": 2093, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "15_carlos_0_65_65_retarget", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/beat2_english/g1/15_carlos_0_65_65_retarget.npz", "split": "test", "text_path": null, "video_summary": "giving a speech", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00132.npz", "sample_index": 132}
{"caption": "", "dataset": "beat2_original_humanref_test", "fps": 30.0, "frames": 60, "global_index": 165061, "index": 94, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00133.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": null, "segment_action": "giving a speech", "segment_frame_end": 2093, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "15_carlos_0_65_65_retarget", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/beat2_english/g1/15_carlos_0_65_65_retarget.npz", "split": "test", "text_path": null, "video_summary": "giving a speech", "window_end": 1076, "window_start": 1016}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00133.npz", "sample_index": 133}
{"caption": "", "dataset": "beat2_original_humanref_test", "fps": 30.0, "frames": 60, "global_index": 165066, "index": 99, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00134.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": null, "segment_action": "giving a speech", "segment_frame_end": 1313, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "12_zhao_2_101_101_retarget", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/beat2_chinese/g1/12_zhao_2_101_101_retarget.npz", "split": "test", "text_path": null, "video_summary": "giving a speech", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00134.npz", "sample_index": 134}
{"caption": "", "dataset": "beat2_original_humanref_test", "fps": 30.0, "frames": 60, "global_index": 165068, "index": 101, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00135.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": null, "segment_action": "giving a speech", "segment_frame_end": 1313, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "12_zhao_2_101_101_retarget", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/beat2_chinese/g1/12_zhao_2_101_101_retarget.npz", "split": "test", "text_path": null, "video_summary": "giving a speech", "window_end": 1313, "window_start": 1253}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00135.npz", "sample_index": 135}
{"caption": "", "dataset": "beat2_original_humanref_test", "fps": 30.0, "frames": 60, "global_index": 165072, "index": 105, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00136.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": null, "segment_action": "giving a speech", "segment_frame_end": 1043, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "12_zhao_2_107_107_retarget", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/beat2_chinese/g1/12_zhao_2_107_107_retarget.npz", "split": "test", "text_path": null, "video_summary": "giving a speech", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00136.npz", "sample_index": 136}
{"caption": "", "dataset": "beat2_original_humanref_test", "fps": 30.0, "frames": 60, "global_index": 165076, "index": 109, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00137.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": null, "segment_action": "giving a speech", "segment_frame_end": 1883, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "12_zhao_2_22_22_retarget", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/beat2_chinese/g1/12_zhao_2_22_22_retarget.npz", "split": "test", "text_path": null, "video_summary": "giving a speech", "window_end": 972, "window_start": 912}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00137.npz", "sample_index": 137}
{"caption": "", "dataset": "beat2_original_humanref_test", "fps": 30.0, "frames": 60, "global_index": 165080, "index": 113, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00138.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": null, "segment_action": "giving a speech", "segment_frame_end": 1493, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "12_zhao_2_39_39_retarget", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/beat2_chinese/g1/12_zhao_2_39_39_retarget.npz", "split": "test", "text_path": null, "video_summary": "giving a speech", "window_end": 1493, "window_start": 1433}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00138.npz", "sample_index": 138}
{"caption": "", "dataset": "beat2_original_humanref_test", "fps": 30.0, "frames": 60, "global_index": 165084, "index": 117, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00139.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": null, "segment_action": "giving a speech", "segment_frame_end": 803, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "12_zhao_2_55_55_retarget", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/beat2_chinese/g1/12_zhao_2_55_55_retarget.npz", "split": "test", "text_path": null, "video_summary": "giving a speech", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00139.npz", "sample_index": 139}
{"caption": "", "dataset": "beat2_original_humanref_test", "fps": 30.0, "frames": 60, "global_index": 165089, "index": 122, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00140.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": null, "segment_action": "giving a speech", "segment_frame_end": 1043, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "12_zhao_2_83_83_retarget", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/beat2_chinese/g1/12_zhao_2_83_83_retarget.npz", "split": "test", "text_path": null, "video_summary": "giving a speech", "window_end": 1043, "window_start": 983}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00140.npz", "sample_index": 140}
{"caption": "", "dataset": "finedance_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 163220, "index": 17, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00141.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/labels/007_retarget__seg0005_000600_000840.json", "segment_action": "a dancer executes fluid arm gestures, steps, lunges, and coordinated body twists", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "graceful expressive", "sequence_name": "007_retarget__seg0005_000600_000840", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/g1/007_retarget__seg0005_000600_000840.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a graceful expressive fine dance with fluid arm movements and coordinated steps", "window_end": 240, "window_start": 180}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00141.npz", "sample_index": 141}
{"caption": "", "dataset": "finedance_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 163247, "index": 44, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00142.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/labels/007_retarget__seg0014_001680_001920.json", "segment_action": "a dancer waves arms, twists torso, steps, bends knees, and kneels to the ground", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "expressive dynamic", "sequence_name": "007_retarget__seg0014_001680_001920", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/g1/007_retarget__seg0014_001680_001920.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs an expressive dynamic sequence with varied arm movements, torso twists, steps, and a kneel", "window_end": 240, "window_start": 180}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00142.npz", "sample_index": 142}
{"caption": "", "dataset": "finedance_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 163267, "index": 64, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00143.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/labels/007_retarget__seg0021_002520_002760.json", "segment_action": "a dancer steps side to side, swings arms, twists torso, and shifts weight rhythmically", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "groovy rhythmic", "sequence_name": "007_retarget__seg0021_002520_002760", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/g1/007_retarget__seg0021_002520_002760.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a groovy rhythmic sequence with side steps, arm swings, and torso twists", "window_end": 150, "window_start": 90}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00143.npz", "sample_index": 143}
{"caption": "", "dataset": "finedance_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 163275, "index": 72, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00144.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/labels/007_retarget__seg0024_002880_003120.json", "segment_action": "a dancer kneels, bends forward, and executes varied arm sweeps, raises, and torso shifts", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "expressive grounded", "sequence_name": "007_retarget__seg0024_002880_003120", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/g1/007_retarget__seg0024_002880_003120.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs an expressive grounded sequence with kneeling and dynamic arm gestures", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00144.npz", "sample_index": 144}
{"caption": "", "dataset": "finedance_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 163277, "index": 74, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00145.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/labels/007_retarget__seg0024_002880_003120.json", "segment_action": "a dancer kneels, bends forward, and executes varied arm sweeps, raises, and torso shifts", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "expressive grounded", "sequence_name": "007_retarget__seg0024_002880_003120", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/g1/007_retarget__seg0024_002880_003120.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs an expressive grounded sequence with kneeling and dynamic arm gestures", "window_end": 240, "window_start": 180}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00145.npz", "sample_index": 145}
{"caption": "", "dataset": "finedance_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 163307, "index": 104, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00146.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/labels/008_retarget__seg0004_000480_000720.json", "segment_action": "a dancer bends knees, sweeps arms, steps side to side, and shifts torso rhythmically", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "coordinated rhythmic", "sequence_name": "008_retarget__seg0004_000480_000720", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/g1/008_retarget__seg0004_000480_000720.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a coordinated rhythmic sequence with knee bends, arm sweeps, and side steps", "window_end": 240, "window_start": 180}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00146.npz", "sample_index": 146}
{"caption": "", "dataset": "finedance_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 163340, "index": 137, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00147.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/labels/008_retarget__seg0015_001800_002040.json", "segment_action": "a dancer raises arms, twists torso, steps side to side, turns, and uses dynamic hand gestures", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "energetic expressive", "sequence_name": "008_retarget__seg0015_001800_002040", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/g1/008_retarget__seg0015_001800_002040.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs an energetic expressive dance with arm gestures, torso twists, steps, and turns", "window_end": 240, "window_start": 180}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00147.npz", "sample_index": 147}
{"caption": "", "dataset": "finedance_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 163344, "index": 141, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00148.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/labels/008_retarget__seg0017_002040_002280.json", "segment_action": "a dancer moves arms fluidly, twists torso, shifts weight, and steps rhythmically", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "graceful expressive", "sequence_name": "008_retarget__seg0017_002040_002280", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/g1/008_retarget__seg0017_002040_002280.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a graceful expressive dance with fluid arm gestures, torso twists, and rhythmic footwork", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00148.npz", "sample_index": 148}
{"caption": "", "dataset": "finedance_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 163380, "index": 177, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00149.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/labels/008_retarget__seg0029_003480_003720.json", "segment_action": "a dancer spins, raises arms in waves, sways hips, and steps rhythmically with weight shifts", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "graceful expressive", "sequence_name": "008_retarget__seg0029_003480_003720", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/g1/008_retarget__seg0029_003480_003720.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a graceful expressive sequence with spins, arm gestures, hip sways, and rhythmic stepping", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00149.npz", "sample_index": 149}
{"caption": "", "dataset": "finedance_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 163394, "index": 191, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00150.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/labels/012_retarget__seg0001_000120_000360.json", "segment_action": "a dancer raises arms, twists torso, sways hips, bends knees, and makes fluid gestures", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "graceful expressive", "sequence_name": "012_retarget__seg0001_000120_000360", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/g1/012_retarget__seg0001_000120_000360.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a graceful expressive dance with fluid arm movements, torso twists, and hip sways", "window_end": 240, "window_start": 180}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00150.npz", "sample_index": 150}
{"caption": "", "dataset": "finedance_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 163400, "index": 197, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00151.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/labels/012_retarget__seg0003_000360_000600.json", "segment_action": "a dancer raises arms, twists torso, steps side to side, bends knees, and makes rhythmic gestures", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "expressive coordinated", "sequence_name": "012_retarget__seg0003_000360_000600", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/g1/012_retarget__seg0003_000360_000600.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs an expressive coordinated sequence with arm movements, torso twists, steps, and knee bends", "window_end": 240, "window_start": 180}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00151.npz", "sample_index": 151}
{"caption": "", "dataset": "finedance_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 163425, "index": 222, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00152.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/labels/012_retarget__seg0012_001440_001680.json", "segment_action": "a dancer waves arms, twists torso, steps side to side, and shifts weight with knee bends", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "expressive coordinated", "sequence_name": "012_retarget__seg0012_001440_001680", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/g1/012_retarget__seg0012_001440_001680.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs an expressive coordinated sequence with arm gestures, torso bends, and rhythmic side steps", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00152.npz", "sample_index": 152}
{"caption": "", "dataset": "finedance_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 163454, "index": 251, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00153.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/labels/012_retarget__seg0021_002520_002760.json", "segment_action": "a dancer swings arms, twists torso, sways hips, steps side to side with rhythmic knee bends", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "groovy rhythmic", "sequence_name": "012_retarget__seg0021_002520_002760", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/g1/012_retarget__seg0021_002520_002760.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a groovy rhythmic dance with arm swings, torso twists, hip sways, and side steps", "window_end": 240, "window_start": 180}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00153.npz", "sample_index": 153}
{"caption": "", "dataset": "finedance_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 163601, "index": 398, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00154.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/labels/014_retarget__seg0047_005640_005880.json", "segment_action": "a dancer bends knees, waves arms, twists torso, and shifts weight in coordinated steps", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "groovy rhythmic", "sequence_name": "014_retarget__seg0047_005640_005880", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/g1/014_retarget__seg0047_005640_005880.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a groovy rhythmic dance with knee bends, arm gestures, torso twists, and weight shifts", "window_end": 240, "window_start": 180}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00154.npz", "sample_index": 154}
{"caption": "", "dataset": "finedance_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 163654, "index": 451, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00155.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/labels/019_retarget__seg0009_001080_001320.json", "segment_action": "a dancer spins, waves arms, drops to knees, steps side to side, twists torso", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "dynamic expressive", "sequence_name": "019_retarget__seg0009_001080_001320", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/g1/019_retarget__seg0009_001080_001320.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs dynamic expressive dance with spins, arm gestures, knee bends, and rhythmic steps", "window_end": 150, "window_start": 90}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00155.npz", "sample_index": 155}
{"caption": "", "dataset": "finedance_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 163681, "index": 478, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00156.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/labels/019_retarget__seg0019_002280_002520.json", "segment_action": "a dancer twists torso, turns body, bends knees, kneels, and takes rhythmic steps", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "expressive dynamic", "sequence_name": "019_retarget__seg0019_002280_002520", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/g1/019_retarget__seg0019_002280_002520.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs an expressive dynamic sequence with body turns, knee bends, kneeling, and rhythmic steps", "window_end": 150, "window_start": 90}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00156.npz", "sample_index": 156}
{"caption": "", "dataset": "finedance_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 163756, "index": 553, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00157.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/labels/019_retarget__seg0047_005640_005880.json", "segment_action": "a dancer waves arms, sways hips, steps side to side, and shifts weight rhythmically", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "groovy rhythmic", "sequence_name": "019_retarget__seg0047_005640_005880", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/g1/019_retarget__seg0047_005640_005880.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a groovy rhythmic dance with arm gestures, hip sways, and side steps", "window_end": 150, "window_start": 90}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00157.npz", "sample_index": 157}
{"caption": "", "dataset": "finedance_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 163837, "index": 634, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00158.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/labels/040_retarget__seg0002_000240_000480.json", "segment_action": "a dancer sweeps arms, twists torso, steps side to side, and spins with coordinated limbs", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "graceful expressive", "sequence_name": "040_retarget__seg0002_000240_000480", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/g1/040_retarget__seg0002_000240_000480.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a graceful expressive sequence with arm sweeps, torso twists, steps, and spins", "window_end": 150, "window_start": 90}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00158.npz", "sample_index": 158}
{"caption": "", "dataset": "finedance_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 163838, "index": 635, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00159.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/labels/040_retarget__seg0002_000240_000480.json", "segment_action": "a dancer sweeps arms, twists torso, steps side to side, and spins with coordinated limbs", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "graceful expressive", "sequence_name": "040_retarget__seg0002_000240_000480", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/g1/040_retarget__seg0002_000240_000480.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a graceful expressive sequence with arm sweeps, torso twists, steps, and spins", "window_end": 240, "window_start": 180}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00159.npz", "sample_index": 159}
{"caption": "", "dataset": "finedance_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 163873, "index": 670, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00160.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/labels/051_retarget__seg0004_000480_000720.json", "segment_action": "a dancer bends knees, raises arms, twists torso, shifts weight, and makes fluid arm gestures", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "graceful expressive", "sequence_name": "051_retarget__seg0004_000480_000720", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/g1/051_retarget__seg0004_000480_000720.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a graceful expressive dance with fluid arm movements, torso twists, and weight shifts", "window_end": 150, "window_start": 90}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00160.npz", "sample_index": 160}
{"caption": "", "dataset": "finedance_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 163905, "index": 702, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00161.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/labels/051_retarget__seg0017_002040_002280.json", "segment_action": "a dancer waves arms, twists torso, spins, kicks legs, and shifts weight dynamically", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "expressive fluid", "sequence_name": "051_retarget__seg0017_002040_002280", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/g1/051_retarget__seg0017_002040_002280.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs an expressive fluid dance with arm gestures, spins, kicks, and torso twists", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00161.npz", "sample_index": 161}
{"caption": "", "dataset": "finedance_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 163919, "index": 716, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00162.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/labels/051_retarget__seg0022_002640_002880.json", "segment_action": "a dancer does arm waves, leg lifts, torso twists, and rhythmic weight shifts with coordination", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "expressive dynamic", "sequence_name": "051_retarget__seg0022_002640_002880", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/g1/051_retarget__seg0022_002640_002880.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs an expressive dynamic fine dance with arm gestures, leg lifts, and coordinated steps", "window_end": 240, "window_start": 180}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00162.npz", "sample_index": 162}
{"caption": "", "dataset": "finedance_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 163924, "index": 721, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00163.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/labels/051_retarget__seg0024_002880_003120.json", "segment_action": "a dancer makes varied arm gestures, bends torso, squats knees, and shifts weight rhythmically", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "expressive dynamic", "sequence_name": "051_retarget__seg0024_002880_003120", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/g1/051_retarget__seg0024_002880_003120.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs an expressive dynamic dance with arm gestures, torso bends, and rhythmic weight shifts", "window_end": 150, "window_start": 90}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00163.npz", "sample_index": 163}
{"caption": "", "dataset": "finedance_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 163960, "index": 757, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00164.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/labels/066_retarget__seg0007_000840_001080.json", "segment_action": "a dancer steps side to side, waves arms, sways hips, twists torso, and uses dynamic hand gestures", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "groovy expressive", "sequence_name": "066_retarget__seg0007_000840_001080", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/g1/066_retarget__seg0007_000840_001080.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a groovy expressive sequence with arm movements, side steps, hip sways, and torso twists", "window_end": 150, "window_start": 90}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00164.npz", "sample_index": 164}
{"caption": "", "dataset": "finedance_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 163962, "index": 759, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00165.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/labels/066_retarget__seg0008_000960_001200.json", "segment_action": "a dancer executes varied arm gestures, sways hips, bends knees, and shifts weight with rhythmic steps", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "groovy rhythmic", "sequence_name": "066_retarget__seg0008_000960_001200", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/g1/066_retarget__seg0008_000960_001200.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a groovy rhythmic dance with dynamic arm gestures, hip sways, and coordinated footwork", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00165.npz", "sample_index": 165}
{"caption": "", "dataset": "finedance_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 164041, "index": 838, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00166.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/labels/071_retarget__seg0023_002760_003000.json", "segment_action": "a dancer waves arms, sways hips, bends knees, and steps side to side rhythmically", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "groovy rhythmic", "sequence_name": "071_retarget__seg0023_002760_003000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/g1/071_retarget__seg0023_002760_003000.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a groovy rhythmic dance with arm gestures, knee bends, and coordinated steps", "window_end": 150, "window_start": 90}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00166.npz", "sample_index": 166}
{"caption": "", "dataset": "finedance_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 164065, "index": 862, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00167.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/labels/071_retarget__seg0031_003720_003960.json", "segment_action": "a dancer makes dynamic arm gestures, twists torso, steps side to side, and shifts weight", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "groovy rhythmic", "sequence_name": "071_retarget__seg0031_003720_003960", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/g1/071_retarget__seg0031_003720_003960.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a groovy rhythmic sequence with arm gestures, torso twists, and side steps", "window_end": 150, "window_start": 90}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00167.npz", "sample_index": 167}
{"caption": "", "dataset": "finedance_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 164118, "index": 915, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00168.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/labels/100_retarget__seg0003_000360_000600.json", "segment_action": "a dancer twists torso, sways hips, moves arms fluidly, and steps with rhythmic weight shifts", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "groovy rhythmic", "sequence_name": "100_retarget__seg0003_000360_000600", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/g1/100_retarget__seg0003_000360_000600.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a groovy rhythmic dance with fluid arm movements, torso twists, and rhythmic footwork", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00168.npz", "sample_index": 168}
{"caption": "", "dataset": "finedance_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 164153, "index": 950, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00169.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/labels/100_retarget__seg0014_001680_001920.json", "segment_action": "a dancer uses varied arm gestures, twists torso, bends knees, and steps rhythmically", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "expressive rhythmic", "sequence_name": "100_retarget__seg0014_001680_001920", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/g1/100_retarget__seg0014_001680_001920.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs an expressive rhythmic sequence with arm gestures, torso twists, and coordinated steps", "window_end": 240, "window_start": 180}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00169.npz", "sample_index": 169}
{"caption": "", "dataset": "finedance_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 164225, "index": 1022, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00170.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/labels/100_retarget__seg0038_004560_004800.json", "segment_action": "a dancer moves arms fluidly, twists torso, sways hips, and steps side to side with weight shifts", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "groovy rhythmic", "sequence_name": "100_retarget__seg0038_004560_004800", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/g1/100_retarget__seg0038_004560_004800.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a groovy rhythmic dance with fluid arm gestures, torso twists, hip sways, and side steps", "window_end": 240, "window_start": 180}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00170.npz", "sample_index": 170}
{"caption": "", "dataset": "finedance_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 164231, "index": 1028, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00171.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/labels/100_retarget__seg0040_004800_005040.json", "segment_action": "a dancer sways hips, twists torso, moves arms, bends knees, and shifts weight rhythmically", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "groovy rhythmic", "sequence_name": "100_retarget__seg0040_004800_005040", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/g1/100_retarget__seg0040_004800_005040.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a groovy rhythmic sequence with hip sways, torso twists, arm gestures, and rhythmic steps", "window_end": 240, "window_start": 180}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00171.npz", "sample_index": 171}
{"caption": "", "dataset": "finedance_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 164313, "index": 1110, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00172.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/labels/108_retarget__seg0009_001080_001320.json", "segment_action": "a dancer extends arms, twists torso, shifts weight, and executes smooth coordinated turns", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "graceful expressive", "sequence_name": "108_retarget__seg0009_001080_001320", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/g1/108_retarget__seg0009_001080_001320.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a graceful expressive sequence with fluid arm movements, torso twists, and weight shifts", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00172.npz", "sample_index": 172}
{"caption": "", "dataset": "finedance_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 164342, "index": 1139, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00173.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/labels/108_retarget__seg0019_002280_002520.json", "segment_action": "a dancer waves arms, twists torso, bends knees, steps side to side, and shifts weight", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "expressive coordinated", "sequence_name": "108_retarget__seg0019_002280_002520", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/g1/108_retarget__seg0019_002280_002520.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs an expressive, coordinated dance with arm gestures, torso twists, and rhythmic steps", "window_end": 240, "window_start": 180}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00173.npz", "sample_index": 173}
{"caption": "", "dataset": "finedance_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 164432, "index": 1229, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00174.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/labels/113_retarget__seg0028_003360_003600.json", "segment_action": "a dancer spins, extends arms in varied poses, steps rhythmically, and shifts weight smoothly", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "graceful expressive", "sequence_name": "113_retarget__seg0028_003360_003600", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/g1/113_retarget__seg0028_003360_003600.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a graceful expressive fine dance with spins, arm poses, and coordinated steps", "window_end": 240, "window_start": 180}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00174.npz", "sample_index": 174}
{"caption": "", "dataset": "finedance_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 164573, "index": 1370, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00175.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/labels/143_retarget__seg0015_001800_002040.json", "segment_action": "a dancer waves arms, spins, shifts weight, and executes coordinated footwork", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "graceful expressive", "sequence_name": "143_retarget__seg0015_001800_002040", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/g1/143_retarget__seg0015_001800_002040.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a graceful expressive dance with flowing arm gestures, spins, and coordinated footwork", "window_end": 240, "window_start": 180}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00175.npz", "sample_index": 175}
{"caption": "", "dataset": "finedance_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 164580, "index": 1377, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00176.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/labels/143_retarget__seg0018_002160_002400.json", "segment_action": "a dancer bends torso, raises arms, lifts legs, spins, and shifts weight rhythmically", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "expressive fluid", "sequence_name": "143_retarget__seg0018_002160_002400", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/g1/143_retarget__seg0018_002160_002400.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs an expressive fluid dance sequence with torso bends, arm gestures, spins, and leg lifts", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00176.npz", "sample_index": 176}
{"caption": "", "dataset": "finedance_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 164617, "index": 1414, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00177.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/labels/143_retarget__seg0030_003600_003840.json", "segment_action": "a dancer raises arms, spins, steps side to side, leans torso, and makes rhythmic gestures", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "expressive coordinated", "sequence_name": "143_retarget__seg0030_003600_003840", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/g1/143_retarget__seg0030_003600_003840.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs an expressive coordinated fine dance with arm movements, spins, steps, and torso leans", "window_end": 150, "window_start": 90}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00177.npz", "sample_index": 177}
{"caption": "", "dataset": "finedance_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 164629, "index": 1426, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00178.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/labels/143_retarget__seg0034_004080_004320.json", "segment_action": "a dancer raises an arm overhead, bends knees, steps back, and lowers the arm smoothly", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "graceful expressive", "sequence_name": "143_retarget__seg0034_004080_004320", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/g1/143_retarget__seg0034_004080_004320.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a graceful expressive sequence with arm raises, knee bends, and backward steps", "window_end": 150, "window_start": 90}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00178.npz", "sample_index": 178}
{"caption": "", "dataset": "finedance_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 164645, "index": 1442, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00179.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/labels/143_retarget__seg0039_004680_004920.json", "segment_action": "a dancer raises arms, sweeps them wide, sways torso, and bends knees fluidly", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "graceful expressive", "sequence_name": "143_retarget__seg0039_004680_004920", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/g1/143_retarget__seg0039_004680_004920.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a graceful expressive sequence with fluid arm movements, torso sways, and knee bends", "window_end": 240, "window_start": 180}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00179.npz", "sample_index": 179}
{"caption": "", "dataset": "finedance_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 164665, "index": 1462, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00180.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/labels/143_retarget__seg0046_005520_005760.json", "segment_action": "a dancer takes deliberate forward steps, shifts weight smoothly, arms relaxed at sides", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "slow controlled", "sequence_name": "143_retarget__seg0046_005520_005760", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/g1/143_retarget__seg0046_005520_005760.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a slow, controlled sequence with deliberate forward steps", "window_end": 150, "window_start": 90}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00180.npz", "sample_index": 180}
{"caption": "", "dataset": "finedance_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 164696, "index": 1493, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00181.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/labels/162_retarget__seg0009_001080_001320.json", "segment_action": "A dancer waves arms, sways hips, bends knees, and shifts weight with torso twists", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "groovy rhythmic", "sequence_name": "162_retarget__seg0009_001080_001320", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/g1/162_retarget__seg0009_001080_001320.npz", "split": "test", "text_path": null, "video_summary": "A dancer performs a groovy rhythmic sequence with arm gestures, hip sways, and knee bends", "window_end": 240, "window_start": 180}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00181.npz", "sample_index": 181}
{"caption": "", "dataset": "finedance_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 164721, "index": 1518, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00182.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/labels/162_retarget__seg0018_002160_002400.json", "segment_action": "a dancer bends knees, steps side to side, twists torso, moves arms, and spins", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "groovy coordinated", "sequence_name": "162_retarget__seg0018_002160_002400", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/g1/162_retarget__seg0018_002160_002400.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a groovy coordinated sequence with knee bends, side steps, arm gestures, and spins", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00182.npz", "sample_index": 182}
{"caption": "", "dataset": "finedance_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 164777, "index": 1574, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00183.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/labels/171_retarget__seg0005_000600_000840.json", "segment_action": "a dancer raises arms, twists torso, sways hips, turns, and uses dynamic hand gestures with knee bends", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "expressive coordinated", "sequence_name": "171_retarget__seg0005_000600_000840", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/g1/171_retarget__seg0005_000600_000840.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs an expressive coordinated dance with arm gestures, torso twists, turns, and hip sways", "window_end": 240, "window_start": 180}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00183.npz", "sample_index": 183}
{"caption": "", "dataset": "finedance_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 164831, "index": 1628, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00184.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/labels/184_retarget__seg0010_001200_001440.json", "segment_action": "a dancer lifts knees, waves arms, bends torso, steps side to side, and ends with a bow", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "playful expressive", "sequence_name": "184_retarget__seg0010_001200_001440", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/g1/184_retarget__seg0010_001200_001440.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a playful expressive sequence with knee lifts, arm gestures, torso bends, and a final bow", "window_end": 240, "window_start": 180}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00184.npz", "sample_index": 184}
{"caption": "", "dataset": "finedance_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 164832, "index": 1629, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00185.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/labels/184_retarget__seg0011_001320_001560.json", "segment_action": "a dancer steps side to side, waves arms, bends torso, and makes coordinated hand gestures", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "playful rhythmic", "sequence_name": "184_retarget__seg0011_001320_001560", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/g1/184_retarget__seg0011_001320_001560.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a playful rhythmic sequence with side steps, arm gestures, and torso bends", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00185.npz", "sample_index": 185}
{"caption": "", "dataset": "finedance_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 164856, "index": 1653, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00186.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/labels/184_retarget__seg0019_002280_002520.json", "segment_action": "a dancer makes fluid arm gestures, twists torso, turns, and steps with rhythmic shifts", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "expressive rhythmic", "sequence_name": "184_retarget__seg0019_002280_002520", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/g1/184_retarget__seg0019_002280_002520.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs an expressive rhythmic sequence with arm gestures, turns, and coordinated stepping", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00186.npz", "sample_index": 186}
{"caption": "", "dataset": "finedance_original_audio_test", "fps": 30.0, "frames": 60, "global_index": 164865, "index": 1662, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00187.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/labels/184_retarget__seg0025_003000_003240.json", "segment_action": "a dancer waves arms, gestures hands, bends knees, twists torso, and shifts weight rhythmically", "segment_frame_end": 240, "segment_frame_start": 0, "segment_index": 0, "segment_style": "playful expressive", "sequence_name": "184_retarget__seg0025_003000_003240", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/finedance/g1/184_retarget__seg0025_003000_003240.npz", "split": "test", "text_path": null, "video_summary": "a dancer performs a playful expressive dance with dynamic arm gestures, knee bends, and rhythmic steps", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00187.npz", "sample_index": 187}
{"caption": "She performs ballet with a strong sense of revulsion.", "dataset": "fitness_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 8985, "index": 0, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00188.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/labels/000004_retarget__seg0000__part0000.json", "segment_action": "She performs ballet with a strong sense of revulsion.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "000004_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/g1/000004_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "She performs ballet with a strong sense of revulsion.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00188.npz", "sample_index": 188}
{"caption": "She performs ballet with a strong sense of revulsion.", "dataset": "fitness_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 8987, "index": 2, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00189.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/labels/000004_retarget__seg0000__part0000.json", "segment_action": "She performs ballet with a strong sense of revulsion.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "000004_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/g1/000004_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "She performs ballet with a strong sense of revulsion.", "window_end": 300, "window_start": 240}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00189.npz", "sample_index": 189}
{"caption": "She performs ballet with a strong sense of revulsion.", "dataset": "fitness_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 8991, "index": 6, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00190.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/labels/000004_retarget__seg0000__part0002.json", "segment_action": "She performs ballet with a strong sense of revulsion.", "segment_frame_end": 236, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "000004_retarget__seg0000__part0002", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/g1/000004_retarget__seg0000__part0002.npz", "split": "test", "text_path": null, "video_summary": "She performs ballet with a strong sense of revulsion.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00190.npz", "sample_index": 190}
{"caption": "She performs ballet with a strong sense of revulsion.", "dataset": "fitness_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 8992, "index": 7, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00191.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/labels/000004_retarget__seg0000__part0002.json", "segment_action": "She performs ballet with a strong sense of revulsion.", "segment_frame_end": 236, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "000004_retarget__seg0000__part0002", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/g1/000004_retarget__seg0000__part0002.npz", "split": "test", "text_path": null, "video_summary": "She performs ballet with a strong sense of revulsion.", "window_end": 148, "window_start": 88}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00191.npz", "sample_index": 191}
{"caption": "She hesitantly performs a ballet.", "dataset": "fitness_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 8994, "index": 9, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00192.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/labels/000015_retarget__seg0000__part0000.json", "segment_action": "She hesitantly performs a ballet.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "000015_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/g1/000015_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "She hesitantly performs a ballet.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00192.npz", "sample_index": 192}
{"caption": "She hesitantly performs a ballet.", "dataset": "fitness_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 8998, "index": 13, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00193.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/labels/000015_retarget__seg0000__part0001.json", "segment_action": "She hesitantly performs a ballet.", "segment_frame_end": 275, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "000015_retarget__seg0000__part0001", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/g1/000015_retarget__seg0000__part0001.npz", "split": "test", "text_path": null, "video_summary": "She hesitantly performs a ballet.", "window_end": 168, "window_start": 108}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00193.npz", "sample_index": 193}
{"caption": "She dances ballet with joy.", "dataset": "fitness_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 9002, "index": 17, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00194.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/labels/000051_retarget__seg0000__part0000.json", "segment_action": "She dances ballet with joy.", "segment_frame_end": 143, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "000051_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/g1/000051_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "She dances ballet with joy.", "window_end": 143, "window_start": 83}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00194.npz", "sample_index": 194}
{"caption": "Concentrating on fitness, jump knee to elbow in a sport.", "dataset": "fitness_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 9003, "index": 18, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00195.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/labels/000060_retarget__seg0000__part0000.json", "segment_action": "Concentrating on fitness, jump knee to elbow in a sport.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "000060_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/g1/000060_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "Concentrating on fitness, jump knee to elbow in a sport.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00195.npz", "sample_index": 195}
{"caption": "Concentrating on fitness, jump knee to elbow in a sport.", "dataset": "fitness_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 9008, "index": 23, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00196.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/labels/000060_retarget__seg0000__part0001.json", "segment_action": "Concentrating on fitness, jump knee to elbow in a sport.", "segment_frame_end": 101, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "000060_retarget__seg0000__part0001", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/g1/000060_retarget__seg0000__part0001.npz", "split": "test", "text_path": null, "video_summary": "Concentrating on fitness, jump knee to elbow in a sport.", "window_end": 101, "window_start": 41}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00196.npz", "sample_index": 196}
{"caption": "Sport fitness involves vigorous jumping motions.", "dataset": "fitness_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 9009, "index": 24, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00197.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/labels/000072_retarget__seg0000__part0000.json", "segment_action": "Sport fitness involves vigorous jumping motions.", "segment_frame_end": 133, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "000072_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/g1/000072_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "Sport fitness involves vigorous jumping motions.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00197.npz", "sample_index": 197}
{"caption": "Runners sprint back and forth, fully focused.", "dataset": "fitness_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 9014, "index": 29, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00198.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/labels/000086_retarget__seg0000__part0000.json", "segment_action": "Runners sprint back and forth, fully focused.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "000086_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/g1/000086_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "Runners sprint back and forth, fully focused.", "window_end": 300, "window_start": 240}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00198.npz", "sample_index": 198}
{"caption": "Runners sprint back and forth, fully focused.", "dataset": "fitness_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 9019, "index": 34, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00199.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/labels/000086_retarget__seg0000__part0002.json", "segment_action": "Runners sprint back and forth, fully focused.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "000086_retarget__seg0000__part0002", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/g1/000086_retarget__seg0000__part0002.npz", "split": "test", "text_path": null, "video_summary": "Runners sprint back and forth, fully focused.", "window_end": 180, "window_start": 120}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00199.npz", "sample_index": 199}
{"caption": "Sport fitness enthusiasts engage in left and right arm lifts while standing.", "dataset": "fitness_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 9021, "index": 36, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00200.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/labels/000100_retarget__seg0000__part0000.json", "segment_action": "Sport fitness enthusiasts engage in left and right arm lifts while standing.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "000100_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/g1/000100_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "Sport fitness enthusiasts engage in left and right arm lifts while standing.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00200.npz", "sample_index": 200}
{"caption": "Sport fitness enthusiasts engage in left and right arm lifts while standing.", "dataset": "fitness_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 9022, "index": 37, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00201.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/labels/000100_retarget__seg0000__part0000.json", "segment_action": "Sport fitness enthusiasts engage in left and right arm lifts while standing.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "000100_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/g1/000100_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "Sport fitness enthusiasts engage in left and right arm lifts while standing.", "window_end": 180, "window_start": 120}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00201.npz", "sample_index": 201}
{"caption": "Sport fitness enthusiasts engage in left and right arm lifts while standing.", "dataset": "fitness_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 9023, "index": 38, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00202.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/labels/000100_retarget__seg0000__part0000.json", "segment_action": "Sport fitness enthusiasts engage in left and right arm lifts while standing.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "000100_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/g1/000100_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "Sport fitness enthusiasts engage in left and right arm lifts while standing.", "window_end": 300, "window_start": 240}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00202.npz", "sample_index": 202}
{"caption": "Sport fitness enthusiasts engage in left and right arm lifts while standing.", "dataset": "fitness_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 9024, "index": 39, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00203.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/labels/000100_retarget__seg0000__part0001.json", "segment_action": "Sport fitness enthusiasts engage in left and right arm lifts while standing.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "000100_retarget__seg0000__part0001", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/g1/000100_retarget__seg0000__part0001.npz", "split": "test", "text_path": null, "video_summary": "Sport fitness enthusiasts engage in left and right arm lifts while standing.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00203.npz", "sample_index": 203}
{"caption": "Sport fitness enthusiasts engage in left and right arm lifts while standing.", "dataset": "fitness_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 9025, "index": 40, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00204.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/labels/000100_retarget__seg0000__part0001.json", "segment_action": "Sport fitness enthusiasts engage in left and right arm lifts while standing.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "000100_retarget__seg0000__part0001", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/g1/000100_retarget__seg0000__part0001.npz", "split": "test", "text_path": null, "video_summary": "Sport fitness enthusiasts engage in left and right arm lifts while standing.", "window_end": 180, "window_start": 120}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00204.npz", "sample_index": 204}
{"caption": "Sport fitness enthusiasts engage in left and right arm lifts while standing.", "dataset": "fitness_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 9027, "index": 42, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00205.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/labels/000100_retarget__seg0000__part0002.json", "segment_action": "Sport fitness enthusiasts engage in left and right arm lifts while standing.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "000100_retarget__seg0000__part0002", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/g1/000100_retarget__seg0000__part0002.npz", "split": "test", "text_path": null, "video_summary": "Sport fitness enthusiasts engage in left and right arm lifts while standing.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00205.npz", "sample_index": 205}
{"caption": "Sport fitness enthusiasts engage in left and right arm lifts while standing.", "dataset": "fitness_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 9030, "index": 45, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00206.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/labels/000100_retarget__seg0000__part0003.json", "segment_action": "Sport fitness enthusiasts engage in left and right arm lifts while standing.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "000100_retarget__seg0000__part0003", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/g1/000100_retarget__seg0000__part0003.npz", "split": "test", "text_path": null, "video_summary": "Sport fitness enthusiasts engage in left and right arm lifts while standing.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00206.npz", "sample_index": 206}
{"caption": "Sport fitness enthusiasts engage in left and right arm lifts while standing.", "dataset": "fitness_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 9031, "index": 46, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00207.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/labels/000100_retarget__seg0000__part0003.json", "segment_action": "Sport fitness enthusiasts engage in left and right arm lifts while standing.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "000100_retarget__seg0000__part0003", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/g1/000100_retarget__seg0000__part0003.npz", "split": "test", "text_path": null, "video_summary": "Sport fitness enthusiasts engage in left and right arm lifts while standing.", "window_end": 180, "window_start": 120}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00207.npz", "sample_index": 207}
{"caption": "Sport fitness enthusiasts engage in left and right arm lifts while standing.", "dataset": "fitness_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 9033, "index": 48, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00208.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/labels/000100_retarget__seg0000__part0004.json", "segment_action": "Sport fitness enthusiasts engage in left and right arm lifts while standing.", "segment_frame_end": 169, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "000100_retarget__seg0000__part0004", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/g1/000100_retarget__seg0000__part0004.npz", "split": "test", "text_path": null, "video_summary": "Sport fitness enthusiasts engage in left and right arm lifts while standing.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00208.npz", "sample_index": 208}
{"caption": "Sport fitness involves lifting arms left and right while standing.", "dataset": "fitness_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 9037, "index": 52, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00209.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/labels/000104_retarget__seg0000__part0000.json", "segment_action": "Sport fitness involves lifting arms left and right while standing.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "000104_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/g1/000104_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "Sport fitness involves lifting arms left and right while standing.", "window_end": 180, "window_start": 120}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00209.npz", "sample_index": 209}
{"caption": "Sport fitness involves lifting arms left and right while standing.", "dataset": "fitness_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 9038, "index": 53, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00210.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/labels/000104_retarget__seg0000__part0000.json", "segment_action": "Sport fitness involves lifting arms left and right while standing.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "000104_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/g1/000104_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "Sport fitness involves lifting arms left and right while standing.", "window_end": 300, "window_start": 240}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00210.npz", "sample_index": 210}
{"caption": "Sport fitness involves lifting arms left and right while standing.", "dataset": "fitness_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 9040, "index": 55, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00211.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/labels/000104_retarget__seg0000__part0001.json", "segment_action": "Sport fitness involves lifting arms left and right while standing.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "000104_retarget__seg0000__part0001", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/g1/000104_retarget__seg0000__part0001.npz", "split": "test", "text_path": null, "video_summary": "Sport fitness involves lifting arms left and right while standing.", "window_end": 180, "window_start": 120}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00211.npz", "sample_index": 211}
{"caption": "Sport fitness involves lifting arms left and right while standing.", "dataset": "fitness_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 9041, "index": 56, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00212.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/labels/000104_retarget__seg0000__part0001.json", "segment_action": "Sport fitness involves lifting arms left and right while standing.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "000104_retarget__seg0000__part0001", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/g1/000104_retarget__seg0000__part0001.npz", "split": "test", "text_path": null, "video_summary": "Sport fitness involves lifting arms left and right while standing.", "window_end": 300, "window_start": 240}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00212.npz", "sample_index": 212}
{"caption": "Sport fitness involves lifting arms left and right while standing.", "dataset": "fitness_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 9042, "index": 57, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00213.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/labels/000104_retarget__seg0000__part0002.json", "segment_action": "Sport fitness involves lifting arms left and right while standing.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "000104_retarget__seg0000__part0002", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/g1/000104_retarget__seg0000__part0002.npz", "split": "test", "text_path": null, "video_summary": "Sport fitness involves lifting arms left and right while standing.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00213.npz", "sample_index": 213}
{"caption": "Sport fitness involves lifting arms left and right while standing.", "dataset": "fitness_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 9044, "index": 59, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00214.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/labels/000104_retarget__seg0000__part0002.json", "segment_action": "Sport fitness involves lifting arms left and right while standing.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "000104_retarget__seg0000__part0002", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/g1/000104_retarget__seg0000__part0002.npz", "split": "test", "text_path": null, "video_summary": "Sport fitness involves lifting arms left and right while standing.", "window_end": 300, "window_start": 240}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00214.npz", "sample_index": 214}
{"caption": "Sport fitness involves lifting arms left and right while standing.", "dataset": "fitness_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 9045, "index": 60, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00215.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/labels/000104_retarget__seg0000__part0003.json", "segment_action": "Sport fitness involves lifting arms left and right while standing.", "segment_frame_end": 274, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "000104_retarget__seg0000__part0003", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/g1/000104_retarget__seg0000__part0003.npz", "split": "test", "text_path": null, "video_summary": "Sport fitness involves lifting arms left and right while standing.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00215.npz", "sample_index": 215}
{"caption": "Standing left and right arm lift is a part of sport fitness, sadly.", "dataset": "fitness_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 9048, "index": 63, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00216.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/labels/000107_retarget__seg0000__part0000.json", "segment_action": "Standing left and right arm lift is a part of sport fitness, sadly.", "segment_frame_end": 74, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "000107_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/g1/000107_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "Standing left and right arm lift is a part of sport fitness, sadly.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00216.npz", "sample_index": 216}
{"caption": "Standing left and right arm lift is a part of sport fitness, sadly.", "dataset": "fitness_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 9049, "index": 64, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00217.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/labels/000107_retarget__seg0000__part0000.json", "segment_action": "Standing left and right arm lift is a part of sport fitness, sadly.", "segment_frame_end": 74, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "000107_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/g1/000107_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "Standing left and right arm lift is a part of sport fitness, sadly.", "window_end": 67, "window_start": 7}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00217.npz", "sample_index": 217}
{"caption": "Standing left and right arm lift is a part of sport fitness, sadly.", "dataset": "fitness_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 9050, "index": 65, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00218.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/labels/000107_retarget__seg0000__part0000.json", "segment_action": "Standing left and right arm lift is a part of sport fitness, sadly.", "segment_frame_end": 74, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "000107_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/g1/000107_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "Standing left and right arm lift is a part of sport fitness, sadly.", "window_end": 74, "window_start": 14}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00218.npz", "sample_index": 218}
{"caption": "Happily, sport fitness enthusiasts perform a standing left and right arm stretch.", "dataset": "fitness_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 9057, "index": 72, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00219.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/labels/000112_retarget__seg0000__part0002.json", "segment_action": "Happily, sport fitness enthusiasts perform a standing left and right arm stretch.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "000112_retarget__seg0000__part0002", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/g1/000112_retarget__seg0000__part0002.npz", "split": "test", "text_path": null, "video_summary": "Happily, sport fitness enthusiasts perform a standing left and right arm stretch.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00219.npz", "sample_index": 219}
{"caption": "Happily, sport fitness enthusiasts perform a standing left and right arm stretch.", "dataset": "fitness_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 9058, "index": 73, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00220.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/labels/000112_retarget__seg0000__part0002.json", "segment_action": "Happily, sport fitness enthusiasts perform a standing left and right arm stretch.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "000112_retarget__seg0000__part0002", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/g1/000112_retarget__seg0000__part0002.npz", "split": "test", "text_path": null, "video_summary": "Happily, sport fitness enthusiasts perform a standing left and right arm stretch.", "window_end": 180, "window_start": 120}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00220.npz", "sample_index": 220}
{"caption": "Happily, sport fitness enthusiasts perform a standing left and right arm stretch.", "dataset": "fitness_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 9059, "index": 74, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00221.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/labels/000112_retarget__seg0000__part0002.json", "segment_action": "Happily, sport fitness enthusiasts perform a standing left and right arm stretch.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "000112_retarget__seg0000__part0002", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/g1/000112_retarget__seg0000__part0002.npz", "split": "test", "text_path": null, "video_summary": "Happily, sport fitness enthusiasts perform a standing left and right arm stretch.", "window_end": 300, "window_start": 240}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00221.npz", "sample_index": 221}
{"caption": "Happily, sport fitness enthusiasts perform a standing left and right arm stretch.", "dataset": "fitness_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 9062, "index": 77, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00222.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/labels/000112_retarget__seg0000__part0003.json", "segment_action": "Happily, sport fitness enthusiasts perform a standing left and right arm stretch.", "segment_frame_end": 79, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "000112_retarget__seg0000__part0003", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/g1/000112_retarget__seg0000__part0003.npz", "split": "test", "text_path": null, "video_summary": "Happily, sport fitness enthusiasts perform a standing left and right arm stretch.", "window_end": 79, "window_start": 19}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00222.npz", "sample_index": 222}
{"caption": "Alternating leg stretches are part of a sport fitness routine.", "dataset": "fitness_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 9070, "index": 85, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00223.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/labels/000129_retarget__seg0000__part0000.json", "segment_action": "Alternating leg stretches are part of a sport fitness routine.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "000129_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/g1/000129_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "Alternating leg stretches are part of a sport fitness routine.", "window_end": 180, "window_start": 120}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00223.npz", "sample_index": 223}
{"caption": "Alternating leg stretches are part of a sport fitness routine.", "dataset": "fitness_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 9071, "index": 86, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00224.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/labels/000129_retarget__seg0000__part0000.json", "segment_action": "Alternating leg stretches are part of a sport fitness routine.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "000129_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/g1/000129_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "Alternating leg stretches are part of a sport fitness routine.", "window_end": 300, "window_start": 240}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00224.npz", "sample_index": 224}
{"caption": "Alternating leg stretches are part of a sport fitness routine.", "dataset": "fitness_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 9072, "index": 87, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00225.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/labels/000129_retarget__seg0000__part0001.json", "segment_action": "Alternating leg stretches are part of a sport fitness routine.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "000129_retarget__seg0000__part0001", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/g1/000129_retarget__seg0000__part0001.npz", "split": "test", "text_path": null, "video_summary": "Alternating leg stretches are part of a sport fitness routine.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00225.npz", "sample_index": 225}
{"caption": "Alternating leg stretches are part of a sport fitness routine.", "dataset": "fitness_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 9073, "index": 88, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00226.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/labels/000129_retarget__seg0000__part0001.json", "segment_action": "Alternating leg stretches are part of a sport fitness routine.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "000129_retarget__seg0000__part0001", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/g1/000129_retarget__seg0000__part0001.npz", "split": "test", "text_path": null, "video_summary": "Alternating leg stretches are part of a sport fitness routine.", "window_end": 180, "window_start": 120}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00226.npz", "sample_index": 226}
{"caption": "Alternating leg stretches are part of a sport fitness routine.", "dataset": "fitness_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 9075, "index": 90, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00227.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/labels/000129_retarget__seg0000__part0002.json", "segment_action": "Alternating leg stretches are part of a sport fitness routine.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "000129_retarget__seg0000__part0002", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/g1/000129_retarget__seg0000__part0002.npz", "split": "test", "text_path": null, "video_summary": "Alternating leg stretches are part of a sport fitness routine.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00227.npz", "sample_index": 227}
{"caption": "Alternating leg stretches are part of a sport fitness routine.", "dataset": "fitness_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 9077, "index": 92, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00228.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/labels/000129_retarget__seg0000__part0002.json", "segment_action": "Alternating leg stretches are part of a sport fitness routine.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "000129_retarget__seg0000__part0002", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/g1/000129_retarget__seg0000__part0002.npz", "split": "test", "text_path": null, "video_summary": "Alternating leg stretches are part of a sport fitness routine.", "window_end": 300, "window_start": 240}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00228.npz", "sample_index": 228}
{"caption": "Standing on one leg, the person swings the other leg in opposite directions with ease.", "dataset": "fitness_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 9079, "index": 94, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00229.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/labels/000133_retarget__seg0000__part0000.json", "segment_action": "Standing on one leg, the person swings the other leg in opposite directions with ease.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "000133_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/g1/000133_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "Standing on one leg, the person swings the other leg in opposite directions with ease.", "window_end": 180, "window_start": 120}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00229.npz", "sample_index": 229}
{"caption": "Standing on one leg, the person swings the other leg in opposite directions with ease.", "dataset": "fitness_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 9080, "index": 95, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00230.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/labels/000133_retarget__seg0000__part0000.json", "segment_action": "Standing on one leg, the person swings the other leg in opposite directions with ease.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "000133_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/g1/000133_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "Standing on one leg, the person swings the other leg in opposite directions with ease.", "window_end": 300, "window_start": 240}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00230.npz", "sample_index": 230}
{"caption": "Standing on one leg, the person swings the other leg in opposite directions with ease.", "dataset": "fitness_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 9083, "index": 98, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00231.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/labels/000133_retarget__seg0000__part0001.json", "segment_action": "Standing on one leg, the person swings the other leg in opposite directions with ease.", "segment_frame_end": 232, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "000133_retarget__seg0000__part0001", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/g1/000133_retarget__seg0000__part0001.npz", "split": "test", "text_path": null, "video_summary": "Standing on one leg, the person swings the other leg in opposite directions with ease.", "window_end": 232, "window_start": 172}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00231.npz", "sample_index": 231}
{"caption": "The right leg stretch is a part of sport fitness.", "dataset": "fitness_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 9084, "index": 99, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00232.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/labels/000243_retarget__seg0000__part0000.json", "segment_action": "The right leg stretch is a part of sport fitness.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "000243_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/g1/000243_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The right leg stretch is a part of sport fitness.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00232.npz", "sample_index": 232}
{"caption": "The right leg stretch is a part of sport fitness.", "dataset": "fitness_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 9091, "index": 106, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00233.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/labels/000243_retarget__seg0000__part0002.json", "segment_action": "The right leg stretch is a part of sport fitness.", "segment_frame_end": 113, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "000243_retarget__seg0000__part0002", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/g1/000243_retarget__seg0000__part0002.npz", "split": "test", "text_path": null, "video_summary": "The right leg stretch is a part of sport fitness.", "window_end": 86, "window_start": 26}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00233.npz", "sample_index": 233}
{"caption": "The right leg stretch is a part of sport fitness.", "dataset": "fitness_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 9092, "index": 107, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00234.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/labels/000243_retarget__seg0000__part0002.json", "segment_action": "The right leg stretch is a part of sport fitness.", "segment_frame_end": 113, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "000243_retarget__seg0000__part0002", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/fitness/g1/000243_retarget__seg0000__part0002.npz", "split": "test", "text_path": null, "video_summary": "The right leg stretch is a part of sport fitness.", "window_end": 113, "window_start": 53}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00234.npz", "sample_index": 234}
{"caption": "a person walks forward while slightly favoring the left.", "dataset": "humanml_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 9133, "index": 40, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00235.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/labels/000141_0_retarget__seg0000__part0000.json", "segment_action": "a person walks forward while slightly favoring the left.", "segment_frame_end": 108, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "000141_0_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/g1/000141_0_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "a person walks forward while slightly favoring the left.", "window_end": 84, "window_start": 24}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00235.npz", "sample_index": 235}
{"caption": "the person is moving something around.", "dataset": "humanml_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 9418, "index": 325, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00236.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/labels/001290_0_retarget__seg0000__part0000.json", "segment_action": "the person is moving something around.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "001290_0_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/g1/001290_0_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "the person is moving something around.", "window_end": 180, "window_start": 120}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00236.npz", "sample_index": 236}
{"caption": "losing balance, moving backwards with both feet.", "dataset": "humanml_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 9516, "index": 423, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00237.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/labels/001664_0_retarget__seg0000__part0000.json", "segment_action": "losing balance, moving backwards with both feet.", "segment_frame_end": 291, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "001664_0_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/g1/001664_0_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "losing balance, moving backwards with both feet.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00237.npz", "sample_index": 237}
{"caption": "a man hoists his right arm into the air", "dataset": "humanml_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 9586, "index": 493, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00238.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/labels/001872_0_retarget__seg0000__part0000.json", "segment_action": "a man hoists his right arm into the air", "segment_frame_end": 186, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "001872_0_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/g1/001872_0_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "a man hoists his right arm into the air", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00238.npz", "sample_index": 238}
{"caption": "a person is collecting an item from its right then petting an animal on the left", "dataset": "humanml_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 9731, "index": 638, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00239.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/labels/002414_0_retarget__seg0000__part0000.json", "segment_action": "a person is collecting an item from its right then petting an animal on the left", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "002414_0_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/g1/002414_0_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "a person is collecting an item from its right then petting an animal on the left", "window_end": 180, "window_start": 120}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00239.npz", "sample_index": 239}
{"caption": "man walks slowly forward with arms raised above his head.", "dataset": "humanml_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 9792, "index": 699, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00240.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/labels/002640_0_retarget__seg0000__part0000.json", "segment_action": "man walks slowly forward with arms raised above his head.", "segment_frame_end": 204, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "002640_0_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/g1/002640_0_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "man walks slowly forward with arms raised above his head.", "window_end": 132, "window_start": 72}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00240.npz", "sample_index": 240}
{"caption": "a person takes five steps starting with their right foot and then stops.", "dataset": "humanml_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 9980, "index": 887, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00241.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/labels/003510_0_retarget__seg0000__part0000.json", "segment_action": "a person takes five steps starting with their right foot and then stops.", "segment_frame_end": 225, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "003510_0_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/g1/003510_0_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "a person takes five steps starting with their right foot and then stops.", "window_end": 225, "window_start": 165}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00241.npz", "sample_index": 241}
{"caption": "a person raised the hadns and started to hit with arms", "dataset": "humanml_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 10475, "index": 1382, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00242.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/labels/005415_0_retarget__seg0000__part0000.json", "segment_action": "a person raised the hadns and started to hit with arms", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "005415_0_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/g1/005415_0_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "a person raised the hadns and started to hit with arms", "window_end": 300, "window_start": 240}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00242.npz", "sample_index": 242}
{"caption": "a person uses their right hand and arm to throw an overarm pitch, while supported by the left arm and hand.", "dataset": "humanml_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 10581, "index": 1488, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00243.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/labels/005835_0_retarget__seg0000__part0000.json", "segment_action": "a person uses their right hand and arm to throw an overarm pitch, while supported by the left arm and hand.", "segment_frame_end": 258, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "005835_0_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/g1/005835_0_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "a person uses their right hand and arm to throw an overarm pitch, while supported by the left arm and hand.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00243.npz", "sample_index": 243}
{"caption": "a person takes four steps forward.", "dataset": "humanml_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 10790, "index": 1697, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00244.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/labels/006561_0_retarget__seg0000__part0000.json", "segment_action": "a person takes four steps forward.", "segment_frame_end": 193, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "006561_0_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/g1/006561_0_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "a person takes four steps forward.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00244.npz", "sample_index": 244}
{"caption": "he picks something up then throwed it", "dataset": "humanml_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 11026, "index": 1933, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00245.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/labels/007383_0_retarget__seg0000__part0000.json", "segment_action": "he picks something up then throwed it", "segment_frame_end": 263, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "007383_0_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/g1/007383_0_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "he picks something up then throwed it", "window_end": 263, "window_start": 203}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00245.npz", "sample_index": 245}
{"caption": "a standing person moving from their right to left then moving backwards and forward again", "dataset": "humanml_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 11086, "index": 1993, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00246.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/labels/007534_0_retarget__seg0000__part0000.json", "segment_action": "a standing person moving from their right to left then moving backwards and forward again", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "007534_0_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/g1/007534_0_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "a standing person moving from their right to left then moving backwards and forward again", "window_end": 180, "window_start": 120}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00246.npz", "sample_index": 246}
{"caption": "person checking the time on his right wrist", "dataset": "humanml_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 11563, "index": 2470, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00247.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/labels/009455_0_retarget__seg0000__part0000.json", "segment_action": "person checking the time on his right wrist", "segment_frame_end": 96, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "009455_0_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/g1/009455_0_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "person checking the time on his right wrist", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00247.npz", "sample_index": 247}
{"caption": "the sim appears to step forward & throw an object.", "dataset": "humanml_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 11856, "index": 2763, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00248.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/labels/010777_0_retarget__seg0000__part0000.json", "segment_action": "the sim appears to step forward & throw an object.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "010777_0_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/g1/010777_0_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "the sim appears to step forward & throw an object.", "window_end": 300, "window_start": 240}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00248.npz", "sample_index": 248}
{"caption": "a person standing with his arms by his sides turns to his right and takes four steps, turns 180 degrees and takes five steps, turns 180 degrees again and takes two steps, stops and faces forward.", "dataset": "humanml_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 11991, "index": 2898, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00249.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/labels/011359_0_retarget__seg0000__part0000.json", "segment_action": "a person standing with his arms by his sides turns to his right and takes four steps, turns 180 degrees and takes five steps, turns 180 degrees again and takes two steps, stops and faces forward.", "segment_frame_end": 298, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "011359_0_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/g1/011359_0_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "a person standing with his arms by his sides turns to his right and takes four steps, turns 180 degrees and takes five steps, turns 180 degrees again and takes two steps, stops and faces forward.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00249.npz", "sample_index": 249}
{"caption": "man stands still straight up and down then bends at the knees and takes one big jump forward.", "dataset": "humanml_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 12305, "index": 3212, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00250.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/labels/012519_0_retarget__seg0000__part0000.json", "segment_action": "man stands still straight up and down then bends at the knees and takes one big jump forward.", "segment_frame_end": 114, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "012519_0_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/g1/012519_0_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "man stands still straight up and down then bends at the knees and takes one big jump forward.", "window_end": 87, "window_start": 27}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00250.npz", "sample_index": 250}
{"caption": "the person is doing horizontal jumps back and forth.", "dataset": "humanml_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 12427, "index": 3334, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00251.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/labels/013126_0_retarget__seg0000__part0000.json", "segment_action": "the person is doing horizontal jumps back and forth.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "013126_0_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/g1/013126_0_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "the person is doing horizontal jumps back and forth.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00251.npz", "sample_index": 251}
{"caption": "a person catches a ball with both hands and then throws it with his right hand.", "dataset": "humanml_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 12737, "index": 3644, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00252.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/labels/014339_0_retarget__seg0000__part0000.json", "segment_action": "a person catches a ball with both hands and then throws it with his right hand.", "segment_frame_end": 237, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "014339_0_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/g1/014339_0_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "a person catches a ball with both hands and then throws it with his right hand.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00252.npz", "sample_index": 252}
{"caption": "the person is walking up the stairs and then down the stairs.", "dataset": "humanml_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 12774, "index": 3681, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00253.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/labels/014476_0_retarget__seg0000__part0000.json", "segment_action": "the person is walking up the stairs and then down the stairs.", "segment_frame_end": 291, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "014476_0_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/g1/014476_0_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "the person is walking up the stairs and then down the stairs.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00253.npz", "sample_index": 253}
{"caption": "a person walks straight forward.", "dataset": "humanml_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 12781, "index": 3688, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00254.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/labels/014495_0_retarget__seg0000__part0000.json", "segment_action": "a person walks straight forward.", "segment_frame_end": 139, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "014495_0_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/g1/014495_0_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "a person walks straight forward.", "window_end": 100, "window_start": 40}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00254.npz", "sample_index": 254}
{"caption": "a person throws something with their right hand, while raising their right leg very slightly.", "dataset": "humanml_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 12791, "index": 3698, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00255.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/labels/014519_0_retarget__seg0000__part0000.json", "segment_action": "a person throws something with their right hand, while raising their right leg very slightly.", "segment_frame_end": 236, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "014519_0_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/g1/014519_0_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "a person throws something with their right hand, while raising their right leg very slightly.", "window_end": 236, "window_start": 176}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00255.npz", "sample_index": 255}
{"caption": "a person jumps, spins completely around in a circle an lands.", "dataset": "humanml_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 12905, "index": 3812, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00256.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/labels/M000183_0_retarget__seg0000__part0000.json", "segment_action": "a person jumps, spins completely around in a circle an lands.", "segment_frame_end": 78, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "M000183_0_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/g1/M000183_0_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "a person jumps, spins completely around in a circle an lands.", "window_end": 69, "window_start": 9}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00256.npz", "sample_index": 256}
{"caption": "a person walks forward to the front and then pivots on one foot to turn around, then walks forward towards the back.", "dataset": "humanml_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 12926, "index": 3833, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00257.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/labels/M000243_0_retarget__seg0000__part0000.json", "segment_action": "a person walks forward to the front and then pivots on one foot to turn around, then walks forward towards the back.", "segment_frame_end": 148, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "M000243_0_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/g1/M000243_0_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "a person walks forward to the front and then pivots on one foot to turn around, then walks forward towards the back.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00257.npz", "sample_index": 257}
{"caption": "a person lifts his hands to catch and throw something", "dataset": "humanml_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 12955, "index": 3862, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00258.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/labels/M000381_0_retarget__seg0000__part0000.json", "segment_action": "a person lifts his hands to catch and throw something", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "M000381_0_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/g1/M000381_0_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "a person lifts his hands to catch and throw something", "window_end": 180, "window_start": 120}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00258.npz", "sample_index": 258}
{"caption": "the person is walking slowly.", "dataset": "humanml_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 13048, "index": 3955, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00259.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/labels/M000736_0_retarget__seg0000__part0000.json", "segment_action": "the person is walking slowly.", "segment_frame_end": 162, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "M000736_0_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/g1/M000736_0_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "the person is walking slowly.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00259.npz", "sample_index": 259}
{"caption": "a person walks forwards stops then turns away and walks back", "dataset": "humanml_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 13248, "index": 4155, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00260.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/labels/M001433_0_retarget__seg0000__part0000.json", "segment_action": "a person walks forwards stops then turns away and walks back", "segment_frame_end": 165, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "M001433_0_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/g1/M001433_0_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "a person walks forwards stops then turns away and walks back", "window_end": 112, "window_start": 52}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00260.npz", "sample_index": 260}
{"caption": "a person stands using his right hand to play a guitar moving it up and down.", "dataset": "humanml_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 13471, "index": 4378, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00261.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/labels/M002259_0_retarget__seg0000__part0000.json", "segment_action": "a person stands using his right hand to play a guitar moving it up and down.", "segment_frame_end": 220, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "M002259_0_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/g1/M002259_0_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "a person stands using his right hand to play a guitar moving it up and down.", "window_end": 220, "window_start": 160}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00261.npz", "sample_index": 261}
{"caption": "a person walks forward starting with their left foot.", "dataset": "humanml_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 13666, "index": 4573, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00262.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/labels/M002947_0_retarget__seg0000__part0000.json", "segment_action": "a person walks forward starting with their left foot.", "segment_frame_end": 159, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "M002947_0_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/g1/M002947_0_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "a person walks forward starting with their left foot.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00262.npz", "sample_index": 262}
{"caption": "a person walks slightly forward then turns right and continues to walk.", "dataset": "humanml_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 14000, "index": 4907, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00263.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/labels/M004099_0_retarget__seg0000__part0000.json", "segment_action": "a person walks slightly forward then turns right and continues to walk.", "segment_frame_end": 118, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "M004099_0_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/g1/M004099_0_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "a person walks slightly forward then turns right and continues to walk.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00263.npz", "sample_index": 263}
{"caption": "a man walks in a s-shape.", "dataset": "humanml_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 14062, "index": 4969, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00264.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/labels/M004290_0_retarget__seg0000__part0000.json", "segment_action": "a man walks in a s-shape.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "M004290_0_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/g1/M004290_0_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "a man walks in a s-shape.", "window_end": 300, "window_start": 240}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00264.npz", "sample_index": 264}
{"caption": "a person takes a small jump forward with their arms stationed at their side.", "dataset": "humanml_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 14285, "index": 5192, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00265.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/labels/M005187_0_retarget__seg0000__part0000.json", "segment_action": "a person takes a small jump forward with their arms stationed at their side.", "segment_frame_end": 85, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "M005187_0_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/g1/M005187_0_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "a person takes a small jump forward with their arms stationed at their side.", "window_end": 85, "window_start": 25}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00265.npz", "sample_index": 265}
{"caption": "a person shuffles to the left, the shuffles to the right.", "dataset": "humanml_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 14374, "index": 5281, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00266.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/labels/M005544_0_retarget__seg0000__part0000.json", "segment_action": "a person shuffles to the left, the shuffles to the right.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "M005544_0_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/g1/M005544_0_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "a person shuffles to the left, the shuffles to the right.", "window_end": 180, "window_start": 120}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00266.npz", "sample_index": 266}
{"caption": "the man plays catch and throw a ball.", "dataset": "humanml_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 14651, "index": 5558, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00267.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/labels/M006659_0_retarget__seg0000__part0000.json", "segment_action": "the man plays catch and throw a ball.", "segment_frame_end": 279, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "M006659_0_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/g1/M006659_0_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "the man plays catch and throw a ball.", "window_end": 279, "window_start": 219}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00267.npz", "sample_index": 267}
{"caption": "a person claps their hands together.", "dataset": "humanml_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 14882, "index": 5789, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00268.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/labels/M007558_0_retarget__seg0000__part0000.json", "segment_action": "a person claps their hands together.", "segment_frame_end": 93, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "M007558_0_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/g1/M007558_0_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "a person claps their hands together.", "window_end": 76, "window_start": 16}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00268.npz", "sample_index": 268}
{"caption": "a person brings his hands together then kicks one leg out in front of him", "dataset": "humanml_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 15026, "index": 5933, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00269.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/labels/M008102_0_retarget__seg0000__part0000.json", "segment_action": "a person brings his hands together then kicks one leg out in front of him", "segment_frame_end": 151, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "M008102_0_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/g1/M008102_0_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "a person brings his hands together then kicks one leg out in front of him", "window_end": 151, "window_start": 91}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00269.npz", "sample_index": 269}
{"caption": "a person jumps into the air.", "dataset": "humanml_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 15059, "index": 5966, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00270.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/labels/M008235_0_retarget__seg0000__part0000.json", "segment_action": "a person jumps into the air.", "segment_frame_end": 116, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "M008235_0_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/g1/M008235_0_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "a person jumps into the air.", "window_end": 116, "window_start": 56}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00270.npz", "sample_index": 270}
{"caption": "person is using a sword", "dataset": "humanml_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 15158, "index": 6065, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00271.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/labels/M008601_0_retarget__seg0000__part0000.json", "segment_action": "person is using a sword", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "M008601_0_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/g1/M008601_0_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "person is using a sword", "window_end": 300, "window_start": 240}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00271.npz", "sample_index": 271}
{"caption": "a person walks forward lifting kicking feet up behind him.", "dataset": "humanml_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 15205, "index": 6112, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00272.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/labels/M008823_0_retarget__seg0000__part0000.json", "segment_action": "a person walks forward lifting kicking feet up behind him.", "segment_frame_end": 127, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "M008823_0_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/g1/M008823_0_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "a person walks forward lifting kicking feet up behind him.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00272.npz", "sample_index": 272}
{"caption": "a person is standing with wobbly knees and moving his arms back and fourth as if he was dizzy.", "dataset": "humanml_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 15259, "index": 6166, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00273.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/labels/M008987_0_retarget__seg0000__part0000.json", "segment_action": "a person is standing with wobbly knees and moving his arms back and fourth as if he was dizzy.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "M008987_0_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/g1/M008987_0_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "a person is standing with wobbly knees and moving his arms back and fourth as if he was dizzy.", "window_end": 180, "window_start": 120}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00273.npz", "sample_index": 273}
{"caption": "the figure does circles with his arms stops and then raises his hands parallel to the floor.", "dataset": "humanml_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 15409, "index": 6316, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00274.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/labels/M009542_0_retarget__seg0000__part0000.json", "segment_action": "the figure does circles with his arms stops and then raises his hands parallel to the floor.", "segment_frame_end": 298, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "M009542_0_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/g1/M009542_0_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "the figure does circles with his arms stops and then raises his hands parallel to the floor.", "window_end": 179, "window_start": 119}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00274.npz", "sample_index": 274}
{"caption": "a person walks with one foot in front of the other trying to keep balance.", "dataset": "humanml_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 15429, "index": 6336, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00275.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/labels/M009660_0_retarget__seg0000__part0000.json", "segment_action": "a person walks with one foot in front of the other trying to keep balance.", "segment_frame_end": 180, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "M009660_0_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/g1/M009660_0_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "a person walks with one foot in front of the other trying to keep balance.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00275.npz", "sample_index": 275}
{"caption": "a person frantically waves their right hand around in the air in front of them.", "dataset": "humanml_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 15790, "index": 6697, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00276.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/labels/M011312_0_retarget__seg0000__part0000.json", "segment_action": "a person frantically waves their right hand around in the air in front of them.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "M011312_0_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/g1/M011312_0_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "a person frantically waves their right hand around in the air in front of them.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00276.npz", "sample_index": 276}
{"caption": "person backed up to throw a ball.", "dataset": "humanml_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 16023, "index": 6930, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00277.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/labels/M012169_0_retarget__seg0000__part0000.json", "segment_action": "person backed up to throw a ball.", "segment_frame_end": 229, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "M012169_0_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/g1/M012169_0_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "person backed up to throw a ball.", "window_end": 144, "window_start": 84}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00277.npz", "sample_index": 277}
{"caption": "a person walking forward while pausing periodically.", "dataset": "humanml_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 16313, "index": 7220, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00278.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/labels/M013094_0_retarget__seg0000__part0000.json", "segment_action": "a person walking forward while pausing periodically.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "M013094_0_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/g1/M013094_0_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "a person walking forward while pausing periodically.", "window_end": 300, "window_start": 240}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00278.npz", "sample_index": 278}
{"caption": "a person moves backwards and sits in a chair.", "dataset": "humanml_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 16386, "index": 7293, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00279.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/labels/M013456_0_retarget__seg0000__part0000.json", "segment_action": "a person moves backwards and sits in a chair.", "segment_frame_end": 214, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "M013456_0_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/g1/M013456_0_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "a person moves backwards and sits in a chair.", "window_end": 214, "window_start": 154}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00279.npz", "sample_index": 279}
{"caption": "a person waves enthusiastically with their left hand.", "dataset": "humanml_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 16575, "index": 7482, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00280.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/labels/M014172_0_retarget__seg0000__part0000.json", "segment_action": "a person waves enthusiastically with their left hand.", "segment_frame_end": 121, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "M014172_0_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/g1/M014172_0_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "a person waves enthusiastically with their left hand.", "window_end": 121, "window_start": 61}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00280.npz", "sample_index": 280}
{"caption": "person appears to be directing traffic with both hands.", "dataset": "humanml_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 16641, "index": 7548, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00281.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/labels/M014475_0_retarget__seg0000__part0000.json", "segment_action": "person appears to be directing traffic with both hands.", "segment_frame_end": 211, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "M014475_0_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/humanml/g1/M014475_0_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "person appears to be directing traffic with both hands.", "window_end": 211, "window_start": 151}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00281.npz", "sample_index": 281}
{"caption": "He sits, repeatedly leans forward with hands sliding toward its feet, then returns to an upright seated position.; style: cyclic", "dataset": "idea400_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 16739, "index": 46, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00282.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/labels/Back_Pain_during_sitting_1_clip1_retarget__seg0000__part0000.json", "segment_action": "He sits, repeatedly leans forward with hands sliding toward its feet, then returns to an upright seated position.", "segment_frame_end": 246, "segment_frame_start": 0, "segment_index": 0, "segment_style": "cyclic", "sequence_name": "Back_Pain_during_sitting_1_clip1_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/g1/Back_Pain_during_sitting_1_clip1_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "He sits, repeatedly leans forward with hands sliding toward its feet, then returns to an upright seated position.; style: cyclic", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00282.npz", "sample_index": 282}
{"caption": "The humanoid robot sits and moves its hands repeatedly from behind its head to the front, running them through its hair to simulate drying motions.; style: cyclic", "dataset": "idea400_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 16925, "index": 232, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00283.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/labels/Drying_hair_during_sitting_2_clip1_retarget__seg0000__part0000.json", "segment_action": "The humanoid robot sits and moves its hands repeatedly from behind its head to the front, running them through its hair to simulate drying motions.", "segment_frame_end": 167, "segment_frame_start": 0, "segment_index": 0, "segment_style": "cyclic", "sequence_name": "Drying_hair_during_sitting_2_clip1_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/g1/Drying_hair_during_sitting_2_clip1_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The humanoid robot sits and moves its hands repeatedly from behind its head to the front, running them through its hair to simulate drying motions.; style: cyclic", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00283.npz", "sample_index": 283}
{"caption": "The humanoid robot remains seated, sequentially extends and retracts its right arm while its fingers perform simulated swiping and scrolling gestures.; style: controlled", "dataset": "idea400_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 16987, "index": 294, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00284.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/labels/Finger_scrolling_interface_operation_and_sitting_at_the_same_time_clip1_retarget__seg0000__part0000.json", "segment_action": "The humanoid robot remains seated, sequentially extends and retracts its right arm while its fingers perform simulated swiping and scrolling gestures.", "segment_frame_end": 129, "segment_frame_start": 0, "segment_index": 0, "segment_style": "controlled", "sequence_name": "Finger_scrolling_interface_operation_and_sitting_at_the_same_time_clip1_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/g1/Finger_scrolling_interface_operation_and_sitting_at_the_same_time_clip1_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The humanoid robot remains seated, sequentially extends and retracts its right arm while its fingers perform simulated swiping and scrolling gestures.; style: controlled", "window_end": 129, "window_start": 69}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00284.npz", "sample_index": 284}
{"caption": "A man walks forward, alternately touching its nose with each hand in sync with its steps: as one leg advances, the opposite hand reaches the nose, then the other leg steps forward while the other hand touches the nose.; style: rhythmic", "dataset": "idea400_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 17080, "index": 387, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00285.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/labels/Hand_touches_nose_and_walking_at_the_same_time_2_clip1_retarget__seg0000__part0000.json", "segment_action": "A man walks forward, alternately touching its nose with each hand in sync with its steps: as one leg advances, the opposite hand reaches the nose, then the other leg steps forward while the other hand touches the nose.", "segment_frame_end": 225, "segment_frame_start": 0, "segment_index": 0, "segment_style": "rhythmic", "sequence_name": "Hand_touches_nose_and_walking_at_the_same_time_2_clip1_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/g1/Hand_touches_nose_and_walking_at_the_same_time_2_clip1_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A man walks forward, alternately touching its nose with each hand in sync with its steps: as one leg advances, the opposite hand reaches the nose, then the other leg steps forward while the other hand touches the nose.; style: rhythmic", "window_end": 142, "window_start": 82}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00285.npz", "sample_index": 285}
{"caption": "He sits, rests its left hand on its left thigh, and intermittently scratches its head with its right hand.; style: subtle", "dataset": "idea400_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 17124, "index": 431, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00286.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/labels/Head_scratching_and_sitting_at_the_same_time_1_clip1_retarget__seg0000__part0000.json", "segment_action": "He sits, rests its left hand on its left thigh, and intermittently scratches its head with its right hand.", "segment_frame_end": 159, "segment_frame_start": 0, "segment_index": 0, "segment_style": "subtle", "sequence_name": "Head_scratching_and_sitting_at_the_same_time_1_clip1_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/g1/Head_scratching_and_sitting_at_the_same_time_1_clip1_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "He sits, rests its left hand on its left thigh, and intermittently scratches its head with its right hand.; style: subtle", "window_end": 159, "window_start": 99}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00286.npz", "sample_index": 286}
{"caption": "Someone stands upright, starts with both hands at chest level, extends its right hand outward while curling fingers to display numbers seven through ten, keeps its left hand stationary with palm facing out and fingers spread, then lowers both hands.; style: controlled", "dataset": "idea400_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 17294, "index": 601, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00287.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/labels/Number_12_while_standing_clip1_retarget__seg0000__part0000.json", "segment_action": "Someone stands upright, starts with both hands at chest level, extends its right hand outward while curling fingers to display numbers seven through ten, keeps its left hand stationary with palm facing out and fingers spread, then lowers both hands.", "segment_frame_end": 232, "segment_frame_start": 0, "segment_index": 0, "segment_style": "controlled", "sequence_name": "Number_12_while_standing_clip1_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/g1/Number_12_while_standing_clip1_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "Someone stands upright, starts with both hands at chest level, extends its right hand outward while curling fingers to display numbers seven through ten, keeps its left hand stationary with palm facing out and fingers spread, then lowers both hands.; style: controlled", "window_end": 146, "window_start": 86}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00287.npz", "sample_index": 287}
{"caption": "Sorry, I can't provide the information you're asking for.", "dataset": "idea400_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 17467, "index": 774, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00288.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/labels/Riding_a_Carriage_while_standing_clip1_retarget__seg0000__part0000.json", "segment_action": "Sorry, I can't provide the information you're asking for.", "segment_frame_end": 192, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "Riding_a_Carriage_while_standing_clip1_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/g1/Riding_a_Carriage_while_standing_clip1_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "Sorry, I can't provide the information you're asking for.", "window_end": 126, "window_start": 66}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00288.npz", "sample_index": 288}
{"caption": "The humanoid robot sits, extends its arms upward as if holding a ball, then follows through with its hands as if releasing the ball.; style: controlled", "dataset": "idea400_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 17554, "index": 861, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00289.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/labels/Shooting_the_Basket_during_sitting_clip1_retarget__seg0000__part0000.json", "segment_action": "The humanoid robot sits, extends its arms upward as if holding a ball, then follows through with its hands as if releasing the ball.", "segment_frame_end": 127, "segment_frame_start": 0, "segment_index": 0, "segment_style": "controlled", "sequence_name": "Shooting_the_Basket_during_sitting_clip1_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/g1/Shooting_the_Basket_during_sitting_clip1_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The humanoid robot sits, extends its arms upward as if holding a ball, then follows through with its hands as if releasing the ball.; style: controlled", "window_end": 127, "window_start": 67}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00289.npz", "sample_index": 289}
{"caption": "A man maintains a standing pose, performing small hand gestures near its face without significant full-body motion or locomotion.; style: playful", "dataset": "idea400_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 17572, "index": 879, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00290.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/labels/Simultaneously_Act_cute_and_standing_clip1_retarget__seg0000__part0000.json", "segment_action": "A man maintains a standing pose, performing small hand gestures near its face without significant full-body motion or locomotion.", "segment_frame_end": 140, "segment_frame_start": 0, "segment_index": 0, "segment_style": "playful", "sequence_name": "Simultaneously_Act_cute_and_standing_clip1_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/g1/Simultaneously_Act_cute_and_standing_clip1_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A man maintains a standing pose, performing small hand gestures near its face without significant full-body motion or locomotion.; style: playful", "window_end": 140, "window_start": 80}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00290.npz", "sample_index": 290}
{"caption": "Someone stands still, then raises one hand to shade its eyes, before returning the hand to a neutral position.; style: slow controlled", "dataset": "idea400_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 17600, "index": 907, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00291.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/labels/Simultaneously_Basking_in_the_sun_and_standing_clip1_retarget__seg0000__part0000.json", "segment_action": "Someone stands still, then raises one hand to shade its eyes, before returning the hand to a neutral position.", "segment_frame_end": 144, "segment_frame_start": 0, "segment_index": 0, "segment_style": "slow controlled", "sequence_name": "Simultaneously_Basking_in_the_sun_and_standing_clip1_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/g1/Simultaneously_Basking_in_the_sun_and_standing_clip1_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "Someone stands still, then raises one hand to shade its eyes, before returning the hand to a neutral position.; style: slow controlled", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00291.npz", "sample_index": 291}
{"caption": "A person walks forward, tilting its head downward in unison with each step, while adjusting its arms to maintain balance.; style: coordinated", "dataset": "idea400_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 17618, "index": 925, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00292.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/labels/Simultaneously_Bow_your_head_and_walking_2_clip1_retarget__seg0000__part0000.json", "segment_action": "A person walks forward, tilting its head downward in unison with each step, while adjusting its arms to maintain balance.", "segment_frame_end": 184, "segment_frame_start": 0, "segment_index": 0, "segment_style": "coordinated", "sequence_name": "Simultaneously_Bow_your_head_and_walking_2_clip1_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/g1/Simultaneously_Bow_your_head_and_walking_2_clip1_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A person walks forward, tilting its head downward in unison with each step, while adjusting its arms to maintain balance.; style: coordinated", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00292.npz", "sample_index": 292}
{"caption": "Someone squats and extends its arms as if arranging objects, then stands and alternates leg and arm movements to mimic walking on the spot.; style: playful", "dataset": "idea400_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 17632, "index": 939, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00293.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/labels/Simultaneously_Camping_campfire_and_walking_clip1_retarget__seg0000__part0000.json", "segment_action": "Someone squats and extends its arms as if arranging objects, then stands and alternates leg and arm movements to mimic walking on the spot.", "segment_frame_end": 245, "segment_frame_start": 0, "segment_index": 0, "segment_style": "playful", "sequence_name": "Simultaneously_Camping_campfire_and_walking_clip1_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/g1/Simultaneously_Camping_campfire_and_walking_clip1_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "Someone squats and extends its arms as if arranging objects, then stands and alternates leg and arm movements to mimic walking on the spot.; style: playful", "window_end": 245, "window_start": 185}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00293.npz", "sample_index": 293}
{"caption": "Someone stands stationary, holds imaginary objects, and repeatedly moves its hands together in a chopping motion.; style: controlled", "dataset": "idea400_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 17660, "index": 967, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00294.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/labels/Simultaneously_Cutting_Vegetables_and_standing_1_clip1_retarget__seg0000__part0000.json", "segment_action": "Someone stands stationary, holds imaginary objects, and repeatedly moves its hands together in a chopping motion.", "segment_frame_end": 232, "segment_frame_start": 0, "segment_index": 0, "segment_style": "controlled", "sequence_name": "Simultaneously_Cutting_Vegetables_and_standing_1_clip1_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/g1/Simultaneously_Cutting_Vegetables_and_standing_1_clip1_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "Someone stands stationary, holds imaginary objects, and repeatedly moves its hands together in a chopping motion.; style: controlled", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00294.npz", "sample_index": 294}
{"caption": "Someone stands in place, performing a series of arm movements: raising its arms above its head, then lowering them to different positions at its sides.; style: controlled", "dataset": "idea400_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 17807, "index": 1114, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00295.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/labels/Simultaneously_Piloting_and_standing_clip1_retarget__seg0000__part0000.json", "segment_action": "Someone stands in place, performing a series of arm movements: raising its arms above its head, then lowering them to different positions at its sides.", "segment_frame_end": 176, "segment_frame_start": 0, "segment_index": 0, "segment_style": "controlled", "sequence_name": "Simultaneously_Piloting_and_standing_clip1_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/g1/Simultaneously_Piloting_and_standing_clip1_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "Someone stands in place, performing a series of arm movements: raising its arms above its head, then lowering them to different positions at its sides.; style: controlled", "window_end": 176, "window_start": 116}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00295.npz", "sample_index": 295}
{"caption": "A person bends over by flexing its hips and knees, extends its arms toward its feet, and performs motions to grasp and remove its footwear.; style: controlled", "dataset": "idea400_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 17926, "index": 1233, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00296.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/labels/Simultaneously_Taking_off_shoes_and_sitting_3_clip1_retarget__seg0000__part0001.json", "segment_action": "A person bends over by flexing its hips and knees, extends its arms toward its feet, and performs motions to grasp and remove its footwear.", "segment_frame_end": 96, "segment_frame_start": 0, "segment_index": 0, "segment_style": "controlled", "sequence_name": "Simultaneously_Taking_off_shoes_and_sitting_3_clip1_retarget__seg0000__part0001", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/g1/Simultaneously_Taking_off_shoes_and_sitting_3_clip1_retarget__seg0000__part0001.npz", "split": "test", "text_path": null, "video_summary": "A person bends over by flexing its hips and knees, extends its arms toward its feet, and performs motions to grasp and remove its footwear.; style: controlled", "window_end": 78, "window_start": 18}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00296.npz", "sample_index": 296}
{"caption": "Sorry, I can't assist with that request.", "dataset": "idea400_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 17991, "index": 1298, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00297.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/labels/Simultaneously_sitting_and_Bouquet_clip1_retarget__seg0000__part0000.json", "segment_action": "Sorry, I can't assist with that request.", "segment_frame_end": 149, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "Simultaneously_sitting_and_Bouquet_clip1_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/g1/Simultaneously_sitting_and_Bouquet_clip1_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "Sorry, I can't assist with that request.", "window_end": 149, "window_start": 89}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00297.npz", "sample_index": 297}
{"caption": "He touches its neck while standing up.; style: controlled", "dataset": "idea400_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 18032, "index": 1339, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00298.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/labels/Simultaneously_sitting_and_Hissing_clip1_retarget__seg0000__part0000.json", "segment_action": "He touches its neck while standing up.", "segment_frame_end": 116, "segment_frame_start": 0, "segment_index": 0, "segment_style": "controlled", "sequence_name": "Simultaneously_sitting_and_Hissing_clip1_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/g1/Simultaneously_sitting_and_Hissing_clip1_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "He touches its neck while standing up.; style: controlled", "window_end": 88, "window_start": 28}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00298.npz", "sample_index": 298}
{"caption": "Someone sits on a chair, alternating between moving its arms to maneuver a kite's control bar or string and adjusting its sitting posture for balance.; style: controlled", "dataset": "idea400_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 18046, "index": 1353, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00299.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/labels/Simultaneously_sitting_and_Kite_Flying_1_clip1_retarget__seg0000__part0000.json", "segment_action": "Someone sits on a chair, alternating between moving its arms to maneuver a kite's control bar or string and adjusting its sitting posture for balance.", "segment_frame_end": 133, "segment_frame_start": 0, "segment_index": 0, "segment_style": "controlled", "sequence_name": "Simultaneously_sitting_and_Kite_Flying_1_clip1_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/g1/Simultaneously_sitting_and_Kite_Flying_1_clip1_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "Someone sits on a chair, alternating between moving its arms to maneuver a kite's control bar or string and adjusting its sitting posture for balance.; style: controlled", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00299.npz", "sample_index": 299}
{"caption": "The humanoid robot sits on a stool, using both hands to wrap around its wrist with coordinated wrist rotations and finger flexions.; style: controlled", "dataset": "idea400_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 18120, "index": 1427, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00300.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/labels/Simultaneously_sitting_and_Wrist_wrap_clip1_retarget__seg0000__part0000.json", "segment_action": "The humanoid robot sits on a stool, using both hands to wrap around its wrist with coordinated wrist rotations and finger flexions.", "segment_frame_end": 167, "segment_frame_start": 0, "segment_index": 0, "segment_style": "controlled", "sequence_name": "Simultaneously_sitting_and_Wrist_wrap_clip1_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/g1/Simultaneously_sitting_and_Wrist_wrap_clip1_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The humanoid robot sits on a stool, using both hands to wrap around its wrist with coordinated wrist rotations and finger flexions.; style: controlled", "window_end": 167, "window_start": 107}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00300.npz", "sample_index": 300}
{"caption": "The humanoid robot stands in a relaxed stance, raises arms horizontally to shoulder level, brings arms forward to clasp hands, lifts clasped hands above the head, separates hands and lowers arms back to shoulder level, then returns arms to relaxed position at sides.; style: controlled", "dataset": "idea400_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 18186, "index": 1493, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00301.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/labels/Simultaneously_standing_and_Number_4_clip1_retarget__seg0000__part0000.json", "segment_action": "The humanoid robot stands in a relaxed stance, raises arms horizontally to shoulder level, brings arms forward to clasp hands, lifts clasped hands above the head, separates hands and lowers arms back to shoulder level, then returns arms to relaxed position at sides.", "segment_frame_end": 231, "segment_frame_start": 0, "segment_index": 0, "segment_style": "controlled", "sequence_name": "Simultaneously_standing_and_Number_4_clip1_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/g1/Simultaneously_standing_and_Number_4_clip1_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The humanoid robot stands in a relaxed stance, raises arms horizontally to shoulder level, brings arms forward to clasp hands, lifts clasped hands above the head, separates hands and lowers arms back to shoulder level, then returns arms to relaxed position at sides.; style: controlled", "window_end": 231, "window_start": 171}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00301.npz", "sample_index": 301}
{"caption": "The humanoid robot stands, shifting its weight repeatedly from one leg to the other.; style: subtle cyclic", "dataset": "idea400_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 18195, "index": 1502, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00302.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/labels/Simultaneously_standing_and_Playing_mobile_games_clip1_retarget__seg0000__part0000.json", "segment_action": "The humanoid robot stands, shifting its weight repeatedly from one leg to the other.", "segment_frame_end": 217, "segment_frame_start": 0, "segment_index": 0, "segment_style": "subtle cyclic", "sequence_name": "Simultaneously_standing_and_Playing_mobile_games_clip1_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/g1/Simultaneously_standing_and_Playing_mobile_games_clip1_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The humanoid robot stands, shifting its weight repeatedly from one leg to the other.; style: subtle cyclic", "window_end": 217, "window_start": 157}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00302.npz", "sample_index": 302}
{"caption": "He walks on the spot, shifting weight between legs and alternating lifting feet, while rhythmically raising hands to perform upbeat gestures including thumbs up and clapping.; style: rhythmic energetic", "dataset": "idea400_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 18265, "index": 1572, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00303.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/labels/Simultaneously_walking_and_Cheer_Up_1_clip1_retarget__seg0000__part0000.json", "segment_action": "He walks on the spot, shifting weight between legs and alternating lifting feet, while rhythmically raising hands to perform upbeat gestures including thumbs up and clapping.", "segment_frame_end": 213, "segment_frame_start": 0, "segment_index": 0, "segment_style": "rhythmic energetic", "sequence_name": "Simultaneously_walking_and_Cheer_Up_1_clip1_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/g1/Simultaneously_walking_and_Cheer_Up_1_clip1_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "He walks on the spot, shifting weight between legs and alternating lifting feet, while rhythmically raising hands to perform upbeat gestures including thumbs up and clapping.; style: rhythmic energetic", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00303.npz", "sample_index": 303}
{"caption": "He walks forward, coordinating alternating leg movements with repeated arm motions that raise its hands to comb its hair.; style: rhythmic", "dataset": "idea400_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 18269, "index": 1576, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00304.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/labels/Simultaneously_walking_and_Comb_your_hair_1_clip1_retarget__seg0000__part0000.json", "segment_action": "He walks forward, coordinating alternating leg movements with repeated arm motions that raise its hands to comb its hair.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "rhythmic", "sequence_name": "Simultaneously_walking_and_Comb_your_hair_1_clip1_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/g1/Simultaneously_walking_and_Comb_your_hair_1_clip1_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "He walks forward, coordinating alternating leg movements with repeated arm motions that raise its hands to comb its hair.; style: rhythmic", "window_end": 180, "window_start": 120}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00304.npz", "sample_index": 304}
{"caption": "A man walks forward while alternately extending its arms forward in a pointing gesture, then retracting them back to a neutral position.; style: rhythmic", "dataset": "idea400_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 18288, "index": 1595, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00305.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/labels/Simultaneously_walking_and_Gestures_in_the_Air_clip1_retarget__seg0000__part0000.json", "segment_action": "A man walks forward while alternately extending its arms forward in a pointing gesture, then retracting them back to a neutral position.", "segment_frame_end": 129, "segment_frame_start": 0, "segment_index": 0, "segment_style": "rhythmic", "sequence_name": "Simultaneously_walking_and_Gestures_in_the_Air_clip1_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/g1/Simultaneously_walking_and_Gestures_in_the_Air_clip1_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A man walks forward while alternately extending its arms forward in a pointing gesture, then retracting them back to a neutral position.; style: rhythmic", "window_end": 129, "window_start": 69}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00305.npz", "sample_index": 305}
{"caption": "The humanoid robot stands, shifts weight between legs, bends and extends knees, and simultaneously swings arms slightly up and down in opposition to leg movements.; style: rhythmic", "dataset": "idea400_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 18487, "index": 1794, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00306.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/labels/Swinging_during_standing_1_clip1_retarget__seg0000__part0000.json", "segment_action": "The humanoid robot stands, shifts weight between legs, bends and extends knees, and simultaneously swings arms slightly up and down in opposition to leg movements.", "segment_frame_end": 182, "segment_frame_start": 0, "segment_index": 0, "segment_style": "rhythmic", "sequence_name": "Swinging_during_standing_1_clip1_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/g1/Swinging_during_standing_1_clip1_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The humanoid robot stands, shifts weight between legs, bends and extends knees, and simultaneously swings arms slightly up and down in opposition to leg movements.; style: rhythmic", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00306.npz", "sample_index": 306}
{"caption": "Someone sits and performs traffic direction gestures with its arms, alternating between right and left signals while remaining seated.; style: rhythmic", "dataset": "idea400_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 18541, "index": 1848, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00307.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/labels/Traffic_Direction_during_sitting_clip1_retarget__seg0000__part0000.json", "segment_action": "Someone sits and performs traffic direction gestures with its arms, alternating between right and left signals while remaining seated.", "segment_frame_end": 175, "segment_frame_start": 0, "segment_index": 0, "segment_style": "rhythmic", "sequence_name": "Traffic_Direction_during_sitting_clip1_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/g1/Traffic_Direction_during_sitting_clip1_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "Someone sits and performs traffic direction gestures with its arms, alternating between right and left signals while remaining seated.; style: rhythmic", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00307.npz", "sample_index": 307}
{"caption": "The humanoid robot walks while extending arms laterally with open palms, then alternately moves each arm forward to signal directions, stepping forward and backward to maintain balance.; style: controlled", "dataset": "idea400_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 18544, "index": 1851, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00308.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/labels/Traffic_Direction_while_walking_clip2_retarget__seg0000__part0000.json", "segment_action": "The humanoid robot walks while extending arms laterally with open palms, then alternately moves each arm forward to signal directions, stepping forward and backward to maintain balance.", "segment_frame_end": 105, "segment_frame_start": 0, "segment_index": 0, "segment_style": "controlled", "sequence_name": "Traffic_Direction_while_walking_clip2_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/g1/Traffic_Direction_while_walking_clip2_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The humanoid robot walks while extending arms laterally with open palms, then alternately moves each arm forward to signal directions, stepping forward and backward to maintain balance.; style: controlled", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00308.npz", "sample_index": 308}
{"caption": "The humanoid robot walks while extending arms laterally with open palms, then alternately moves each arm forward to signal directions, stepping forward and backward to maintain balance.; style: controlled", "dataset": "idea400_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 18546, "index": 1853, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00309.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/labels/Traffic_Direction_while_walking_clip2_retarget__seg0000__part0000.json", "segment_action": "The humanoid robot walks while extending arms laterally with open palms, then alternately moves each arm forward to signal directions, stepping forward and backward to maintain balance.", "segment_frame_end": 105, "segment_frame_start": 0, "segment_index": 0, "segment_style": "controlled", "sequence_name": "Traffic_Direction_while_walking_clip2_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/g1/Traffic_Direction_while_walking_clip2_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The humanoid robot walks while extending arms laterally with open palms, then alternately moves each arm forward to signal directions, stepping forward and backward to maintain balance.; style: controlled", "window_end": 105, "window_start": 45}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00309.npz", "sample_index": 309}
{"caption": "A person stands in place, with its arms, hands, torso, and entire body performing subtle, uncontrolled trembling motions continuously.; style: subtle", "dataset": "idea400_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 18551, "index": 1858, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00310.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/labels/Trembling_while_standing_clip1_retarget__seg0000__part0000.json", "segment_action": "A person stands in place, with its arms, hands, torso, and entire body performing subtle, uncontrolled trembling motions continuously.", "segment_frame_end": 135, "segment_frame_start": 0, "segment_index": 0, "segment_style": "subtle", "sequence_name": "Trembling_while_standing_clip1_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/g1/Trembling_while_standing_clip1_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A person stands in place, with its arms, hands, torso, and entire body performing subtle, uncontrolled trembling motions continuously.; style: subtle", "window_end": 98, "window_start": 38}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00310.npz", "sample_index": 310}
{"caption": "The humanoid robot sits stably on a chair, repeatedly clenching one hand into a fist and moving it up and down in a rhythmic motion.; style: rhythmic", "dataset": "idea400_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 18712, "index": 2019, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00311.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/labels/sitting_and_Fist_Shaking_at_the_same_time_3_clip1_retarget__seg0000__part0000.json", "segment_action": "The humanoid robot sits stably on a chair, repeatedly clenching one hand into a fist and moving it up and down in a rhythmic motion.", "segment_frame_end": 145, "segment_frame_start": 0, "segment_index": 0, "segment_style": "rhythmic", "sequence_name": "sitting_and_Fist_Shaking_at_the_same_time_3_clip1_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/g1/sitting_and_Fist_Shaking_at_the_same_time_3_clip1_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The humanoid robot sits stably on a chair, repeatedly clenching one hand into a fist and moving it up and down in a rhythmic motion.; style: rhythmic", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00311.npz", "sample_index": 311}
{"caption": "Someone sits on a chair, repeatedly extends one leg outward, then returns it to its original position, while keeping its torso and head upright and stable throughout the motion.; style: controlled", "dataset": "idea400_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 18738, "index": 2045, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00312.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/labels/sitting_and_Kicking_something_at_the_same_time_1_clip1_retarget__seg0000__part0000.json", "segment_action": "Someone sits on a chair, repeatedly extends one leg outward, then returns it to its original position, while keeping its torso and head upright and stable throughout the motion.", "segment_frame_end": 269, "segment_frame_start": 0, "segment_index": 0, "segment_style": "controlled", "sequence_name": "sitting_and_Kicking_something_at_the_same_time_1_clip1_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/g1/sitting_and_Kicking_something_at_the_same_time_1_clip1_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "Someone sits on a chair, repeatedly extends one leg outward, then returns it to its original position, while keeping its torso and head upright and stable throughout the motion.; style: controlled", "window_end": 269, "window_start": 209}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00312.npz", "sample_index": 312}
{"caption": "The humanoid robot sits on a stool and makes hand gestures while holding a phone to its ear.; style: casual", "dataset": "idea400_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 18856, "index": 2163, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00313.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/labels/sitting_during_Calling_clip1_retarget__seg0000__part0000.json", "segment_action": "The humanoid robot sits on a stool and makes hand gestures while holding a phone to its ear.", "segment_frame_end": 133, "segment_frame_start": 0, "segment_index": 0, "segment_style": "casual", "sequence_name": "sitting_during_Calling_clip1_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/g1/sitting_during_Calling_clip1_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The humanoid robot sits on a stool and makes hand gestures while holding a phone to its ear.; style: casual", "window_end": 96, "window_start": 36}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00313.npz", "sample_index": 313}
{"caption": "He hops on one foot, starting with a slight knee bend, springing off the foot while lifting the other leg, maintaining balance mid-air, then landing on the same foot, with arms aiding balance throughout.; style: energetic", "dataset": "idea400_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 18893, "index": 2200, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00314.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/labels/sitting_during_Hop_on_one_foot_clip1_retarget__seg0000__part0000.json", "segment_action": "He hops on one foot, starting with a slight knee bend, springing off the foot while lifting the other leg, maintaining balance mid-air, then landing on the same foot, with arms aiding balance throughout.", "segment_frame_end": 218, "segment_frame_start": 0, "segment_index": 0, "segment_style": "energetic", "sequence_name": "sitting_during_Hop_on_one_foot_clip1_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/g1/sitting_during_Hop_on_one_foot_clip1_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "He hops on one foot, starting with a slight knee bend, springing off the foot while lifting the other leg, maintaining balance mid-air, then landing on the same foot, with arms aiding balance throughout.; style: energetic", "window_end": 218, "window_start": 158}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00314.npz", "sample_index": 314}
{"caption": "A man sits in a chair, remaining mostly stationary with no significant movement or change in position.; style: static", "dataset": "idea400_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 19050, "index": 2357, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00315.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/labels/sitting_while_Number_8_1_clip1_retarget__seg0000__part0000.json", "segment_action": "A man sits in a chair, remaining mostly stationary with no significant movement or change in position.", "segment_frame_end": 169, "segment_frame_start": 0, "segment_index": 0, "segment_style": "static", "sequence_name": "sitting_while_Number_8_1_clip1_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/g1/sitting_while_Number_8_1_clip1_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A man sits in a chair, remaining mostly stationary with no significant movement or change in position.; style: static", "window_end": 169, "window_start": 109}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00315.npz", "sample_index": 315}
{"caption": "The humanoid robot sits on a stool, repeatedly partially stands up, moves its arms as if using a device, then sits back down.; style: cyclic", "dataset": "idea400_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 19103, "index": 2410, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00316.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/labels/sitting_while_Wearing_a_Dress_clip1_retarget__seg0000__part0000.json", "segment_action": "The humanoid robot sits on a stool, repeatedly partially stands up, moves its arms as if using a device, then sits back down.", "segment_frame_end": 155, "segment_frame_start": 0, "segment_index": 0, "segment_style": "cyclic", "sequence_name": "sitting_while_Wearing_a_Dress_clip1_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/g1/sitting_while_Wearing_a_Dress_clip1_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The humanoid robot sits on a stool, repeatedly partially stands up, moves its arms as if using a device, then sits back down.; style: cyclic", "window_end": 108, "window_start": 48}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00316.npz", "sample_index": 316}
{"caption": "Someone stands in place, repeatedly raises one arm to shoulder height, bends its elbow, and reaches its hand toward the opposite arm's elbow in a sequential motion.; style: cyclic", "dataset": "idea400_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 19299, "index": 2606, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00317.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/labels/standing_during_Hand_touching_elbow_clip1_retarget__seg0000__part0000.json", "segment_action": "Someone stands in place, repeatedly raises one arm to shoulder height, bends its elbow, and reaches its hand toward the opposite arm's elbow in a sequential motion.", "segment_frame_end": 213, "segment_frame_start": 0, "segment_index": 0, "segment_style": "cyclic", "sequence_name": "standing_during_Hand_touching_elbow_clip1_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/g1/standing_during_Hand_touching_elbow_clip1_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "Someone stands in place, repeatedly raises one arm to shoulder height, bends its elbow, and reaches its hand toward the opposite arm's elbow in a sequential motion.; style: cyclic", "window_end": 136, "window_start": 76}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00317.npz", "sample_index": 317}
{"caption": "A man holds a pair of glasses in front of itself with both hands, then positions the glasses onto its head, adjusting them to sit properly while coordinating hand and head movements.; style: controlled", "dataset": "idea400_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 19400, "index": 2707, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00318.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/labels/standing_during_Wearing_Glasses_2_clip1_retarget__seg0000__part0000.json", "segment_action": "A man holds a pair of glasses in front of itself with both hands, then positions the glasses onto its head, adjusting them to sit properly while coordinating hand and head movements.", "segment_frame_end": 176, "segment_frame_start": 0, "segment_index": 0, "segment_style": "controlled", "sequence_name": "standing_during_Wearing_Glasses_2_clip1_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/g1/standing_during_Wearing_Glasses_2_clip1_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A man holds a pair of glasses in front of itself with both hands, then positions the glasses onto its head, adjusting them to sit properly while coordinating hand and head movements.; style: controlled", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00318.npz", "sample_index": 318}
{"caption": "The humanoid robot maintains a stationary upright stance, extends its arms at shoulder height with palms facing down, then performs alternating, cyclic up-and-down arm movements in a repetitive circular pattern.; style: rhythmic cyclic", "dataset": "idea400_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 19471, "index": 2778, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00319.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/labels/standing_while_Merry-Go-Round_1_clip1_retarget__seg0000__part0000.json", "segment_action": "The humanoid robot maintains a stationary upright stance, extends its arms at shoulder height with palms facing down, then performs alternating, cyclic up-and-down arm movements in a repetitive circular pattern.", "segment_frame_end": 169, "segment_frame_start": 0, "segment_index": 0, "segment_style": "rhythmic cyclic", "sequence_name": "standing_while_Merry-Go-Round_1_clip1_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/g1/standing_while_Merry-Go-Round_1_clip1_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The humanoid robot maintains a stationary upright stance, extends its arms at shoulder height with palms facing down, then performs alternating, cyclic up-and-down arm movements in a repetitive circular pattern.; style: rhythmic cyclic", "window_end": 169, "window_start": 109}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00319.npz", "sample_index": 319}
{"caption": "A person stands in place, mimics holding a broom, moves its arms in back-and-forth sweeping motions, and shifts its weight and stance to simulate sweeping.; style: rhythmic", "dataset": "idea400_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 19531, "index": 2838, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00320.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/labels/standing_while_Sweeping_clip1_retarget__seg0000__part0000.json", "segment_action": "A person stands in place, mimics holding a broom, moves its arms in back-and-forth sweeping motions, and shifts its weight and stance to simulate sweeping.", "segment_frame_end": 160, "segment_frame_start": 0, "segment_index": 0, "segment_style": "rhythmic", "sequence_name": "standing_while_Sweeping_clip1_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/g1/standing_while_Sweeping_clip1_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A person stands in place, mimics holding a broom, moves its arms in back-and-forth sweeping motions, and shifts its weight and stance to simulate sweeping.; style: rhythmic", "window_end": 160, "window_start": 100}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00320.npz", "sample_index": 320}
{"caption": "Someone stands, repeatedly performs cyclical arm and hand motions of dipping, scrubbing, and wringing, with slight torso adjustments to support the actions.; style: cyclic", "dataset": "idea400_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 19552, "index": 2859, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00321.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/labels/standing_while_Washing_things_1_clip1_retarget__seg0000__part0000.json", "segment_action": "Someone stands, repeatedly performs cyclical arm and hand motions of dipping, scrubbing, and wringing, with slight torso adjustments to support the actions.", "segment_frame_end": 238, "segment_frame_start": 0, "segment_index": 0, "segment_style": "cyclic", "sequence_name": "standing_while_Washing_things_1_clip1_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/g1/standing_while_Washing_things_1_clip1_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "Someone stands, repeatedly performs cyclical arm and hand motions of dipping, scrubbing, and wringing, with slight torso adjustments to support the actions.; style: cyclic", "window_end": 238, "window_start": 178}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00321.npz", "sample_index": 321}
{"caption": "He walks forward, simultaneously opening its arms wide in a welcoming hug gesture while swinging its arms in coordination with its steps.; style: coordinated", "dataset": "idea400_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 19580, "index": 2887, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00322.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/labels/walking_and_Hug_at_the_same_time_1_clip1_retarget__seg0000__part0000.json", "segment_action": "He walks forward, simultaneously opening its arms wide in a welcoming hug gesture while swinging its arms in coordination with its steps.", "segment_frame_end": 235, "segment_frame_start": 0, "segment_index": 0, "segment_style": "coordinated", "sequence_name": "walking_and_Hug_at_the_same_time_1_clip1_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/g1/walking_and_Hug_at_the_same_time_1_clip1_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "He walks forward, simultaneously opening its arms wide in a welcoming hug gesture while swinging its arms in coordination with its steps.; style: coordinated", "window_end": 235, "window_start": 175}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00322.npz", "sample_index": 322}
{"caption": "The humanoid robot performs an exaggerated forward walk with alternating leg steps, while simultaneously rotating its arms in a freestyle swimming-like motion.; style: energetic", "dataset": "idea400_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 19643, "index": 2950, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00323.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/labels/walking_and_Swimming_at_the_same_time_clip1_retarget__seg0000__part0000.json", "segment_action": "The humanoid robot performs an exaggerated forward walk with alternating leg steps, while simultaneously rotating its arms in a freestyle swimming-like motion.", "segment_frame_end": 218, "segment_frame_start": 0, "segment_index": 0, "segment_style": "energetic", "sequence_name": "walking_and_Swimming_at_the_same_time_clip1_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/g1/walking_and_Swimming_at_the_same_time_clip1_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The humanoid robot performs an exaggerated forward walk with alternating leg steps, while simultaneously rotating its arms in a freestyle swimming-like motion.; style: energetic", "window_end": 218, "window_start": 158}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00323.npz", "sample_index": 323}
{"caption": "A person walks forward, alternating steps with each foot. Then bends down, lifts one foot, uses hands to stretch and adjust a sock over it, then lifts the other foot and repeats the sock-adjusting motion.; style: controlled", "dataset": "idea400_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 19736, "index": 3043, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00324.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/labels/walking_during_Putting_on_socks_clip1_retarget__seg0000__part0000.json", "segment_action": "A person walks forward, alternating steps with each foot. Then bends down, lifts one foot, uses hands to stretch and adjust a sock over it, then lifts the other foot and repeats the sock-adjusting motion.", "segment_frame_end": 211, "segment_frame_start": 0, "segment_index": 0, "segment_style": "controlled", "sequence_name": "walking_during_Putting_on_socks_clip1_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/g1/walking_during_Putting_on_socks_clip1_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A person walks forward, alternating steps with each foot. Then bends down, lifts one foot, uses hands to stretch and adjust a sock over it, then lifts the other foot and repeats the sock-adjusting motion.; style: controlled", "window_end": 136, "window_start": 76}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00324.npz", "sample_index": 324}
{"caption": "A man walks forward, alternating left and right foot steps, and intermittently raises its right hand to scratch its ear before returning the arm to a natural hanging position at its side.; style: controlled", "dataset": "idea400_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 19790, "index": 3097, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00325.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/labels/walking_while_Ear_scratching_clip1_retarget__seg0000__part0000.json", "segment_action": "A man walks forward, alternating left and right foot steps, and intermittently raises its right hand to scratch its ear before returning the arm to a natural hanging position at its side.", "segment_frame_end": 160, "segment_frame_start": 0, "segment_index": 0, "segment_style": "controlled", "sequence_name": "walking_while_Ear_scratching_clip1_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/g1/walking_while_Ear_scratching_clip1_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A man walks forward, alternating left and right foot steps, and intermittently raises its right hand to scratch its ear before returning the arm to a natural hanging position at its side.; style: controlled", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00325.npz", "sample_index": 325}
{"caption": "Someone walks forward, alternately lifting and stepping with each foot, while repeatedly raising its right hand to pat its left shoulder.; style: coordinated", "dataset": "idea400_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 19830, "index": 3137, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00326.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/labels/walking_while_Pat_on_the_shoulder_clip1_retarget__seg0000__part0000.json", "segment_action": "Someone walks forward, alternately lifting and stepping with each foot, while repeatedly raising its right hand to pat its left shoulder.", "segment_frame_end": 155, "segment_frame_start": 0, "segment_index": 0, "segment_style": "coordinated", "sequence_name": "walking_while_Pat_on_the_shoulder_clip1_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/g1/walking_while_Pat_on_the_shoulder_clip1_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "Someone walks forward, alternately lifting and stepping with each foot, while repeatedly raising its right hand to pat its left shoulder.; style: coordinated", "window_end": 108, "window_start": 48}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00326.npz", "sample_index": 326}
{"caption": "A person walks across an area, extends its right arm for a handshake while stepping forward, retracts the arm, then continues walking straight with alternating leg steps.; style: controlled", "dataset": "idea400_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 19846, "index": 3153, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00327.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/labels/walking_while_Shake_hands_and_say_goodbye_clip1_retarget__seg0000__part0000.json", "segment_action": "A person walks across an area, extends its right arm for a handshake while stepping forward, retracts the arm, then continues walking straight with alternating leg steps.", "segment_frame_end": 155, "segment_frame_start": 0, "segment_index": 0, "segment_style": "controlled", "sequence_name": "walking_while_Shake_hands_and_say_goodbye_clip1_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/idea400/g1/walking_while_Shake_hands_and_say_goodbye_clip1_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A person walks across an area, extends its right arm for a handshake while stepping forward, retracts the arm, then continues walking straight with alternating leg steps.; style: controlled", "window_end": 155, "window_start": 95}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00327.npz", "sample_index": 327}
{"caption": "A man stands firmly with arms crossed, occasionally gesturing with his hands as he explains something, displaying a mix of focus and frustration.", "dataset": "motiongv_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 21363, "index": 1501, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00328.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/labels/folder0/005878_retarget__seg0000__part0000.json", "segment_action": "A man stands firmly with arms crossed, occasionally gesturing with his hands as he explains something, displaying a mix of focus and frustration.", "segment_frame_end": 75, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "005878_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/g1/folder0/005878_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A man stands firmly with arms crossed, occasionally gesturing with his hands as he explains something, displaying a mix of focus and frustration.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00328.npz", "sample_index": 328}
{"caption": "A martial artist stands with wide legs and defensive arms, slowly shifting his weight while adjusting his hands and maintaining balance and focus near a body of water.", "dataset": "motiongv_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 22795, "index": 2933, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00329.npz", "meta": {"eval_num_windows": 2, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/labels/folder0/011243_retarget__seg0000__part0000.json", "segment_action": "A martial artist stands with wide legs and defensive arms, slowly shifting his weight while adjusting his hands and maintaining balance and focus near a body of water.", "segment_frame_end": 61, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "011243_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/g1/folder0/011243_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A martial artist stands with wide legs and defensive arms, slowly shifting his weight while adjusting his hands and maintaining balance and focus near a body of water.", "window_end": 61, "window_start": 1}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00329.npz", "sample_index": 329}
{"caption": "A person practices martial arts with focus and calmness.", "dataset": "motiongv_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 24019, "index": 4157, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00330.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/labels/folder0/016259_retarget__seg0000__part0000.json", "segment_action": "A person practices martial arts with focus and calmness.", "segment_frame_end": 135, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "016259_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/g1/folder0/016259_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A person practices martial arts with focus and calmness.", "window_end": 98, "window_start": 38}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00330.npz", "sample_index": 330}
{"caption": "A man in a martial arts uniform stands tall, then assumes a focused stance with feet shoulder-width apart and hands at the center, exuding calm alertness in an indoor dojo.", "dataset": "motiongv_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 28888, "index": 9026, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00331.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/labels/folder0/035683_retarget__seg0000__part0000.json", "segment_action": "A man in a martial arts uniform stands tall, then assumes a focused stance with feet shoulder-width apart and hands at the center, exuding calm alertness in an indoor dojo.", "segment_frame_end": 142, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "035683_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/g1/folder0/035683_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A man in a martial arts uniform stands tall, then assumes a focused stance with feet shoulder-width apart and hands at the center, exuding calm alertness in an indoor dojo.", "window_end": 101, "window_start": 41}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00331.npz", "sample_index": 331}
{"caption": "A person is mastering a martial arts stance.", "dataset": "motiongv_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 34647, "index": 14785, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00332.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/labels/folder1/058162_retarget__seg0000__part0001.json", "segment_action": "A person is mastering a martial arts stance.", "segment_frame_end": 167, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "058162_retarget__seg0000__part0001", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/g1/folder1/058162_retarget__seg0000__part0001.npz", "split": "test", "text_path": null, "video_summary": "A person is mastering a martial arts stance.", "window_end": 114, "window_start": 54}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00332.npz", "sample_index": 332}
{"caption": "A person in a striped shirt performs an energetic dance routine, starting with low bends, rising while moving their hands confidently up and down, striking poses, swaying their hips, and turning in different directions.", "dataset": "motiongv_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 36319, "index": 16457, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00333.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/labels/folder1/064237_retarget__seg0000__part0000.json", "segment_action": "A person in a striped shirt performs an energetic dance routine, starting with low bends, rising while moving their hands confidently up and down, striking poses, swaying their hips, and turning in different directions.", "segment_frame_end": 135, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "064237_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/g1/folder1/064237_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A person in a striped shirt performs an energetic dance routine, starting with low bends, rising while moving their hands confidently up and down, striking poses, swaying their hips, and turning in different directions.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00333.npz", "sample_index": 333}
{"caption": "They execute a movement sequence with arms extended and slightly bent, alternating between left and right.", "dataset": "motiongv_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 39864, "index": 20002, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00334.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/labels/folder1/078392_retarget__seg0000__part0000.json", "segment_action": "They execute a movement sequence with arms extended and slightly bent, alternating between left and right.", "segment_frame_end": 121, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "078392_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/g1/folder1/078392_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "They execute a movement sequence with arms extended and slightly bent, alternating between left and right.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00334.npz", "sample_index": 334}
{"caption": "They perform a lively dance with a repetitive motion.", "dataset": "motiongv_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 41339, "index": 21477, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00335.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/labels/folder1/084585_retarget__seg0000__part0001.json", "segment_action": "They perform a lively dance with a repetitive motion.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "084585_retarget__seg0000__part0001", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/g1/folder1/084585_retarget__seg0000__part0001.npz", "split": "test", "text_path": null, "video_summary": "They perform a lively dance with a repetitive motion.", "window_end": 180, "window_start": 120}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00335.npz", "sample_index": 335}
{"caption": "A man practices martial arts by assuming a low stance with legs apart, raising one hand in a fist, stepping forward, and thrusting with his arm, demonstrating strength and control.", "dataset": "motiongv_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 43549, "index": 23687, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00336.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/labels/folder1/092440_retarget__seg0000__part0000.json", "segment_action": "A man practices martial arts by assuming a low stance with legs apart, raising one hand in a fist, stepping forward, and thrusting with his arm, demonstrating strength and control.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "092440_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/g1/folder1/092440_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A man practices martial arts by assuming a low stance with legs apart, raising one hand in a fist, stepping forward, and thrusting with his arm, demonstrating strength and control.", "window_end": 300, "window_start": 240}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00336.npz", "sample_index": 336}
{"caption": "The person executes a martial arts move, beginning with a low squat, followed by a swift lunge with a striking arm, and finishing with a standing stance.", "dataset": "motiongv_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 44261, "index": 24399, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00337.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/labels/folder1/095158_retarget__seg0000__part0000.json", "segment_action": "The person executes a martial arts move, beginning with a low squat, followed by a swift lunge with a striking arm, and finishing with a standing stance.", "segment_frame_end": 131, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "095158_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/g1/folder1/095158_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The person executes a martial arts move, beginning with a low squat, followed by a swift lunge with a striking arm, and finishing with a standing stance.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00337.npz", "sample_index": 337}
{"caption": "A person in traditional attire executes a fluid sword movement, lifting the sword with both hands and performing a circular motion, exuding a practiced and focused air.", "dataset": "motiongv_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 45093, "index": 25231, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00338.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/labels/folder1/098341_retarget__seg0000__part0000.json", "segment_action": "A person in traditional attire executes a fluid sword movement, lifting the sword with both hands and performing a circular motion, exuding a practiced and focused air.", "segment_frame_end": 95, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "098341_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/g1/folder1/098341_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A person in traditional attire executes a fluid sword movement, lifting the sword with both hands and performing a circular motion, exuding a practiced and focused air.", "window_end": 95, "window_start": 35}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00338.npz", "sample_index": 338}
{"caption": "A performer in traditional attire is executing a fluid martial arts technique.", "dataset": "motiongv_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 47034, "index": 27172, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00339.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/labels/folder1/105005_retarget__seg0000__part0000.json", "segment_action": "A performer in traditional attire is executing a fluid martial arts technique.", "segment_frame_end": 69, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "105005_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/g1/folder1/105005_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A performer in traditional attire is executing a fluid martial arts technique.", "window_end": 69, "window_start": 9}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00339.npz", "sample_index": 339}
{"caption": "A person in loose clothing practices martial arts, stepping forward with their left foot and extending their arms to push against an opponent, while maintaining balance and focus.", "dataset": "motiongv_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 47624, "index": 27762, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00340.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/labels/folder1/107418_retarget__seg0000__part0000.json", "segment_action": "A person in loose clothing practices martial arts, stepping forward with their left foot and extending their arms to push against an opponent, while maintaining balance and focus.", "segment_frame_end": 69, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "107418_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/g1/folder1/107418_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A person in loose clothing practices martial arts, stepping forward with their left foot and extending their arms to push against an opponent, while maintaining balance and focus.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00340.npz", "sample_index": 340}
{"caption": "A woman is doing a boxing workout, starting in a guard position with raised hands.", "dataset": "motiongv_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 49944, "index": 30082, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00341.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/labels/folder2/116174_retarget__seg0000__part0000.json", "segment_action": "A woman is doing a boxing workout, starting in a guard position with raised hands.", "segment_frame_end": 195, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "116174_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/g1/folder2/116174_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A woman is doing a boxing workout, starting in a guard position with raised hands.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00341.npz", "sample_index": 341}
{"caption": "A young man stands still, holding a sign and shifting his weight between his feet, exuding calmness. He adjusts the sign with slight hand movements, conveying focus and engagement through deliberate gestures.", "dataset": "motiongv_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 54668, "index": 34806, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00342.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/labels/folder2/133028_retarget__seg0000__part0001.json", "segment_action": "A young man stands still, holding a sign and shifting his weight between his feet, exuding calmness. He adjusts the sign with slight hand movements, conveying focus and engagement through deliberate gestures.", "segment_frame_end": 123, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "133028_retarget__seg0000__part0001", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/g1/folder2/133028_retarget__seg0000__part0001.npz", "split": "test", "text_path": null, "video_summary": "A young man stands still, holding a sign and shifting his weight between his feet, exuding calmness. He adjusts the sign with slight hand movements, conveying focus and engagement through deliberate gestures.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00342.npz", "sample_index": 342}
{"caption": "A fit woman is jogging in place with a lively rhythm.", "dataset": "motiongv_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 56397, "index": 36535, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00343.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/labels/folder2/138152_retarget__seg0000__part0000.json", "segment_action": "A fit woman is jogging in place with a lively rhythm.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "138152_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/g1/folder2/138152_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A fit woman is jogging in place with a lively rhythm.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00343.npz", "sample_index": 343}
{"caption": "A woman stands still, scanning her surroundings. She uses her right hand to gesture to her left, then to her right, conveying her message with confidence.", "dataset": "motiongv_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 59776, "index": 39914, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00344.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/labels/folder2/150497_retarget__seg0000__part0000.json", "segment_action": "A woman stands still, scanning her surroundings. She uses her right hand to gesture to her left, then to her right, conveying her message with confidence.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "150497_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/g1/folder2/150497_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A woman stands still, scanning her surroundings. She uses her right hand to gesture to her left, then to her right, conveying her message with confidence.", "window_end": 300, "window_start": 240}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00344.npz", "sample_index": 344}
{"caption": "The person remains stationary, briefly adjusting their hair with a fluid, casual motion, conveying a relaxed atmosphere.", "dataset": "motiongv_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 60025, "index": 40163, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00345.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/labels/folder2/151463_retarget__seg0000__part0000.json", "segment_action": "The person remains stationary, briefly adjusting their hair with a fluid, casual motion, conveying a relaxed atmosphere.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "151463_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/g1/folder2/151463_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The person remains stationary, briefly adjusting their hair with a fluid, casual motion, conveying a relaxed atmosphere.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00345.npz", "sample_index": 345}
{"caption": "A young woman dances with great energy.", "dataset": "motiongv_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 64207, "index": 44345, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00346.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/labels/folder3/168954_retarget__seg0000__part0000.json", "segment_action": "A young woman dances with great energy.", "segment_frame_end": 80, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "168954_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/g1/folder3/168954_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A young woman dances with great energy.", "window_end": 80, "window_start": 20}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00346.npz", "sample_index": 346}
{"caption": "A woman executes a choreographed dance. She starts with her arms lowered, then raises one arm while stretching the other outwards. Her body undulates smoothly as she shifts between various arm poses, generating harmonious and rhythmic gestures. Her facial expression conveys delight and poise.", "dataset": "motiongv_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 64548, "index": 44686, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00347.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/labels/folder3/170254_retarget__seg0000__part0000.json", "segment_action": "A woman executes a choreographed dance. She starts with her arms lowered, then raises one arm while stretching the other outwards. Her body undulates smoothly as she shifts between various arm poses, generating harmonious and rhythmic gestures. Her facial expression conveys delight and poise.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "170254_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/g1/folder3/170254_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A woman executes a choreographed dance. She starts with her arms lowered, then raises one arm while stretching the other outwards. Her body undulates smoothly as she shifts between various arm poses, generating harmonious and rhythmic gestures. Her facial expression conveys delight and poise.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00347.npz", "sample_index": 347}
{"caption": "A young man sits engaged, speaking with animated gestures and leaning forward occasionally as he conveys his enthusiasm.", "dataset": "motiongv_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 69821, "index": 49959, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00348.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/labels/folder3/191631_retarget__seg0000__part0000.json", "segment_action": "A young man sits engaged, speaking with animated gestures and leaning forward occasionally as he conveys his enthusiasm.", "segment_frame_end": 99, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "191631_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/g1/folder3/191631_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A young man sits engaged, speaking with animated gestures and leaning forward occasionally as he conveys his enthusiasm.", "window_end": 99, "window_start": 39}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00348.npz", "sample_index": 348}
{"caption": "The person assumes a calm pose, folding their hands in front of their chest, bows slightly with a peaceful expression, conveying respect or prayer. Their movements are deliberate and slow, exuding serenity and intention.", "dataset": "motiongv_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 71211, "index": 51349, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00349.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/labels/folder3/197179_retarget__seg0000__part0000.json", "segment_action": "The person assumes a calm pose, folding their hands in front of their chest, bows slightly with a peaceful expression, conveying respect or prayer. Their movements are deliberate and slow, exuding serenity and intention.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "197179_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/g1/folder3/197179_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The person assumes a calm pose, folding their hands in front of their chest, bows slightly with a peaceful expression, conveying respect or prayer. Their movements are deliberate and slow, exuding serenity and intention.", "window_end": 300, "window_start": 240}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00349.npz", "sample_index": 349}
{"caption": "A young martial artist perfects his technique.", "dataset": "motiongv_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 83042, "index": 63180, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00350.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/labels/folder4/244174_retarget__seg0000__part0000.json", "segment_action": "A young martial artist perfects his technique.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "244174_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/g1/folder4/244174_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A young martial artist perfects his technique.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00350.npz", "sample_index": 350}
{"caption": "A young badminton player stands in the court, holding a racket, and demonstrates a shot with deliberate, enthusiastic movements.", "dataset": "motiongv_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 88573, "index": 68711, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00351.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/labels/folder4/267026_retarget__seg0000__part0000.json", "segment_action": "A young badminton player stands in the court, holding a racket, and demonstrates a shot with deliberate, enthusiastic movements.", "segment_frame_end": 245, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "267026_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/g1/folder4/267026_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A young badminton player stands in the court, holding a racket, and demonstrates a shot with deliberate, enthusiastic movements.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00351.npz", "sample_index": 351}
{"caption": "A woman performs a weighted side squat with a wide stance, alternating arm movements and maintaining a strong, energetic posture.", "dataset": "motiongv_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 88720, "index": 68858, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00352.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/labels/folder4/267674_retarget__seg0000__part0000.json", "segment_action": "A woman performs a weighted side squat with a wide stance, alternating arm movements and maintaining a strong, energetic posture.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "267674_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/g1/folder4/267674_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A woman performs a weighted side squat with a wide stance, alternating arm movements and maintaining a strong, energetic posture.", "window_end": 300, "window_start": 240}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00352.npz", "sample_index": 352}
{"caption": "A skilled player is engaged in a game of billiards, using precise technique to make his shot.", "dataset": "motiongv_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 88837, "index": 68975, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00353.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/labels/folder4/268130_retarget__seg0000__part0001.json", "segment_action": "A skilled player is engaged in a game of billiards, using precise technique to make his shot.", "segment_frame_end": 180, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "268130_retarget__seg0000__part0001", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/g1/folder4/268130_retarget__seg0000__part0001.npz", "split": "test", "text_path": null, "video_summary": "A skilled player is engaged in a game of billiards, using precise technique to make his shot.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00353.npz", "sample_index": 353}
{"caption": "A young man performs a strength exercise by bending down to lift a weight, then stands tall while holding it close to his chest, demonstrating balance and control.", "dataset": "motiongv_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 92274, "index": 72412, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00354.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/labels/folder5/284170_retarget__seg0000__part0001.json", "segment_action": "A young man performs a strength exercise by bending down to lift a weight, then stands tall while holding it close to his chest, demonstrating balance and control.", "segment_frame_end": 77, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "284170_retarget__seg0000__part0001", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/g1/folder5/284170_retarget__seg0000__part0001.npz", "split": "test", "text_path": null, "video_summary": "A young man performs a strength exercise by bending down to lift a weight, then stands tall while holding it close to his chest, demonstrating balance and control.", "window_end": 77, "window_start": 17}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00354.npz", "sample_index": 354}
{"caption": "A woman performs a fluid dance, raising her arms, moving her hands in a flowing motion, and shifting from side to side, occasionally placing her hands on her hips.", "dataset": "motiongv_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 95420, "index": 75558, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00355.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/labels/folder5/298381_retarget__seg0000__part0000.json", "segment_action": "A woman performs a fluid dance, raising her arms, moving her hands in a flowing motion, and shifting from side to side, occasionally placing her hands on her hips.", "segment_frame_end": 103, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "298381_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/g1/folder5/298381_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A woman performs a fluid dance, raising her arms, moving her hands in a flowing motion, and shifting from side to side, occasionally placing her hands on her hips.", "window_end": 82, "window_start": 22}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00355.npz", "sample_index": 355}
{"caption": "A woman performs a lively dance, moving her hips and arms in a carefree manner.", "dataset": "motiongv_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 95647, "index": 75785, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00356.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/labels/folder5/299501_retarget__seg0000__part0000.json", "segment_action": "A woman performs a lively dance, moving her hips and arms in a carefree manner.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "299501_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/g1/folder5/299501_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A woman performs a lively dance, moving her hips and arms in a carefree manner.", "window_end": 180, "window_start": 120}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00356.npz", "sample_index": 356}
{"caption": "A man uses resistance bands for exercise, starting with a bent-knee stance and pulling the bands back with both hands while keeping his back straight.", "dataset": "motiongv_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 96654, "index": 76792, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00357.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/labels/folder5/303766_retarget__seg0000__part0000.json", "segment_action": "A man uses resistance bands for exercise, starting with a bent-knee stance and pulling the bands back with both hands while keeping his back straight.", "segment_frame_end": 95, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "303766_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/g1/folder5/303766_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A man uses resistance bands for exercise, starting with a bent-knee stance and pulling the bands back with both hands while keeping his back straight.", "window_end": 78, "window_start": 18}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00357.npz", "sample_index": 357}
{"caption": "A woman does a lunge exercise, stepping forward with one leg and bending her front knee while keeping her back leg straight, all while maintaining balance with her arms.", "dataset": "motiongv_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 97660, "index": 77798, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00358.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/labels/folder5/308410_retarget__seg0000__part0000.json", "segment_action": "A woman does a lunge exercise, stepping forward with one leg and bending her front knee while keeping her back leg straight, all while maintaining balance with her arms.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "308410_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/g1/folder5/308410_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A woman does a lunge exercise, stepping forward with one leg and bending her front knee while keeping her back leg straight, all while maintaining balance with her arms.", "window_end": 180, "window_start": 120}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00358.npz", "sample_index": 358}
{"caption": "In a loose red outfit, a man executes a fluid martial arts movement, starting with a forward arm raise and single-foot stance.", "dataset": "motiongv_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 99278, "index": 79416, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00359.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/labels/folder5/316011_retarget__seg0000__part0000.json", "segment_action": "In a loose red outfit, a man executes a fluid martial arts movement, starting with a forward arm raise and single-foot stance.", "segment_frame_end": 193, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "316011_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/g1/folder5/316011_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "In a loose red outfit, a man executes a fluid martial arts movement, starting with a forward arm raise and single-foot stance.", "window_end": 126, "window_start": 66}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00359.npz", "sample_index": 359}
{"caption": "A man is honing his boxing skills by executing a series of precise movements.", "dataset": "motiongv_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 99646, "index": 79784, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00360.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/labels/folder5/317631_retarget__seg0000__part0000.json", "segment_action": "A man is honing his boxing skills by executing a series of precise movements.", "segment_frame_end": 81, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "317631_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/g1/folder5/317631_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A man is honing his boxing skills by executing a series of precise movements.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00360.npz", "sample_index": 360}
{"caption": "A person in a red outfit is standing with feet together, then lifts and lowers their arms while shifting weight onto one foot in a controlled manner in an outdoor setting.", "dataset": "motiongv_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 99878, "index": 80016, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00361.npz", "meta": {"eval_num_windows": 2, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/labels/folder5/318675_retarget__seg0000__part0000.json", "segment_action": "A person in a red outfit is standing with feet together, then lifts and lowers their arms while shifting weight onto one foot in a controlled manner in an outdoor setting.", "segment_frame_end": 61, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "318675_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/g1/folder5/318675_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A person in a red outfit is standing with feet together, then lifts and lowers their arms while shifting weight onto one foot in a controlled manner in an outdoor setting.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00361.npz", "sample_index": 361}
{"caption": "A woman sits cross-legged, hands on knees, palms up, maintaining a calm and centered posture.", "dataset": "motiongv_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 102186, "index": 82324, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00362.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/labels/folder5/329215_retarget__seg0000__part0000.json", "segment_action": "A woman sits cross-legged, hands on knees, palms up, maintaining a calm and centered posture.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "329215_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/g1/folder5/329215_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A woman sits cross-legged, hands on knees, palms up, maintaining a calm and centered posture.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00362.npz", "sample_index": 362}
{"caption": "A man stands beside a seated boy in the boxing ring, offering support with a gentle hand gesture.", "dataset": "motiongv_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 103104, "index": 83242, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00363.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/labels/folder5/333501_retarget__seg0000__part0000.json", "segment_action": "A man stands beside a seated boy in the boxing ring, offering support with a gentle hand gesture.", "segment_frame_end": 109, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "333501_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/g1/folder5/333501_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A man stands beside a seated boy in the boxing ring, offering support with a gentle hand gesture.", "window_end": 84, "window_start": 24}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00363.npz", "sample_index": 363}
{"caption": "A woman is exercising vigorously.", "dataset": "motiongv_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 105712, "index": 85850, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00364.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/labels/folder6/346185_retarget__seg0000__part0000.json", "segment_action": "A woman is exercising vigorously.", "segment_frame_end": 99, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "346185_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/g1/folder6/346185_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A woman is exercising vigorously.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00364.npz", "sample_index": 364}
{"caption": "The person is in a yoga pose, lowering their upper body while keeping one leg extended back.", "dataset": "motiongv_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 107786, "index": 87924, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00365.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/labels/folder6/355553_retarget__seg0000__part0000.json", "segment_action": "The person is in a yoga pose, lowering their upper body while keeping one leg extended back.", "segment_frame_end": 259, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "355553_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/g1/folder6/355553_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The person is in a yoga pose, lowering their upper body while keeping one leg extended back.", "window_end": 160, "window_start": 100}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00365.npz", "sample_index": 365}
{"caption": "A woman in a bright pink top conveys her emotions to the camera through expressive gestures and facial expressions.", "dataset": "motiongv_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 108229, "index": 88367, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00366.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/labels/folder6/357382_retarget__seg0000__part0001.json", "segment_action": "A woman in a bright pink top conveys her emotions to the camera through expressive gestures and facial expressions.", "segment_frame_end": 161, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "357382_retarget__seg0000__part0001", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/g1/folder6/357382_retarget__seg0000__part0001.npz", "split": "test", "text_path": null, "video_summary": "A woman in a bright pink top conveys her emotions to the camera through expressive gestures and facial expressions.", "window_end": 161, "window_start": 101}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00366.npz", "sample_index": 366}
{"caption": "A woman performs a stretching exercise on her knees, extending one leg forward and lifting her torso.", "dataset": "motiongv_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 111292, "index": 91430, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00367.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/labels/folder6/371134_retarget__seg0000__part0000.json", "segment_action": "A woman performs a stretching exercise on her knees, extending one leg forward and lifting her torso.", "segment_frame_end": 65, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "371134_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/g1/folder6/371134_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A woman performs a stretching exercise on her knees, extending one leg forward and lifting her torso.", "window_end": 65, "window_start": 5}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00367.npz", "sample_index": 367}
{"caption": "A woman performs a stability ball exercise, starting in a kneeling position and curling her body inward, before returning to the starting position with controlled movements.", "dataset": "motiongv_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 116130, "index": 96268, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00368.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/labels/folder7/392439_retarget__seg0000__part0000.json", "segment_action": "A woman performs a stability ball exercise, starting in a kneeling position and curling her body inward, before returning to the starting position with controlled movements.", "segment_frame_end": 102, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "392439_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/g1/folder7/392439_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A woman performs a stability ball exercise, starting in a kneeling position and curling her body inward, before returning to the starting position with controlled movements.", "window_end": 81, "window_start": 21}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00368.npz", "sample_index": 368}
{"caption": "A person in a blue outfit performs a series of hand movements, standing firm and gesturing outward with his elbows bent, before lowering his arms gradually.", "dataset": "motiongv_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 117073, "index": 97211, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00369.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/labels/folder7/396615_retarget__seg0000__part0000.json", "segment_action": "A person in a blue outfit performs a series of hand movements, standing firm and gesturing outward with his elbows bent, before lowering his arms gradually.", "segment_frame_end": 81, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "396615_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/g1/folder7/396615_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A person in a blue outfit performs a series of hand movements, standing firm and gesturing outward with his elbows bent, before lowering his arms gradually.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00369.npz", "sample_index": 369}
{"caption": "They are engaged in a serious conversation in a formal setting, standing calmly with elegant attire.", "dataset": "motiongv_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 129200, "index": 109338, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00370.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/labels/folder8/451897_retarget__seg0000__part0000.json", "segment_action": "They are engaged in a serious conversation in a formal setting, standing calmly with elegant attire.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "451897_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/g1/folder8/451897_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "They are engaged in a serious conversation in a formal setting, standing calmly with elegant attire.", "window_end": 300, "window_start": 240}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00370.npz", "sample_index": 370}
{"caption": "A person assumes a strong stance, feet apart and knees slightly bent, fists clenched.", "dataset": "motiongv_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 137461, "index": 117599, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00371.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/labels/folder8/491215_retarget__seg0000__part0000.json", "segment_action": "A person assumes a strong stance, feet apart and knees slightly bent, fists clenched.", "segment_frame_end": 283, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "491215_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/g1/folder8/491215_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A person assumes a strong stance, feet apart and knees slightly bent, fists clenched.", "window_end": 283, "window_start": 223}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00371.npz", "sample_index": 371}
{"caption": "The man is doing schlag es weg. A person stands with a shield, assumes a defensive stance, then lifts the shield and sways it back while positioning their other hand for a strike, before transitioning into a new stance, ready to defend with focus and determination.", "dataset": "motiongv_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 139970, "index": 120108, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00372.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/labels/folder9/509173_retarget__seg0000__part0000.json", "segment_action": "The man is doing schlag es weg. A person stands with a shield, assumes a defensive stance, then lifts the shield and sways it back while positioning their other hand for a strike, before transitioning into a new stance, ready to defend with focus and determination.", "segment_frame_end": 62, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "509173_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/g1/folder9/509173_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The man is doing schlag es weg. A person stands with a shield, assumes a defensive stance, then lifts the shield and sways it back while positioning their other hand for a strike, before transitioning into a new stance, ready to defend with focus and determination.", "window_end": 62, "window_start": 2}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00372.npz", "sample_index": 372}
{"caption": "The man is doing Tree pose. The figure in the green outline is doing a tree pose, maintaining balance by standing upright and lifting one arm while placing one foot on the opposite thigh in a calm indoor environment.", "dataset": "motiongv_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 145089, "index": 125227, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00373.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/labels/folder9/552556_retarget__seg0000__part0000.json", "segment_action": "The man is doing Tree pose. The figure in the green outline is doing a tree pose, maintaining balance by standing upright and lifting one arm while placing one foot on the opposite thigh in a calm indoor environment.", "segment_frame_end": 180, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "552556_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motiongv/g1/folder9/552556_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The man is doing Tree pose. The figure in the green outline is doing a tree pose, maintaining balance by standing upright and lifting one arm while placing one foot on the opposite thigh in a calm indoor environment.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00373.npz", "sample_index": 373}
{"caption": "A person is doing dumbbell bicep curls.", "dataset": "motionllama_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 146022, "index": 10, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00374.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/labels/fit3d/train/s03/smplx_slices/dumbbell_biceps_curls_slice_1_retarget__seg0000__part0001.json", "segment_action": "A person is doing dumbbell bicep curls.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "dumbbell_biceps_curls_slice_1_retarget__seg0000__part0001", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/g1/fit3d/train/s03/smplx_slices/dumbbell_biceps_curls_slice_1_retarget__seg0000__part0001.npz", "split": "test", "text_path": null, "video_summary": "A person is doing dumbbell bicep curls.", "window_end": 180, "window_start": 120}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00374.npz", "sample_index": 374}
{"caption": "Start with a right-leg lunge and twist, then a left-leg lunge with a twist.", "dataset": "motionllama_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 146065, "index": 53, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00375.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/labels/fit3d/train/s03/smplx_slices/warmup_16_slice_4_retarget__seg0000__part0001.json", "segment_action": "Start with a right-leg lunge and twist, then a left-leg lunge with a twist.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "warmup_16_slice_4_retarget__seg0000__part0001", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/g1/fit3d/train/s03/smplx_slices/warmup_16_slice_4_retarget__seg0000__part0001.npz", "split": "test", "text_path": null, "video_summary": "Start with a right-leg lunge and twist, then a left-leg lunge with a twist.", "window_end": 300, "window_start": 240}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00375.npz", "sample_index": 375}
{"caption": "The athlete warms up with a lunge stretch, stepping forward with the left leg and raising both arms.", "dataset": "motionllama_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 146069, "index": 57, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00376.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/labels/fit3d/train/s03/smplx_slices/warmup_2_slice_1_retarget__seg0000__part0001.json", "segment_action": "The athlete warms up with a lunge stretch, stepping forward with the left leg and raising both arms.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "warmup_2_slice_1_retarget__seg0000__part0001", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/g1/fit3d/train/s03/smplx_slices/warmup_2_slice_1_retarget__seg0000__part0001.npz", "split": "test", "text_path": null, "video_summary": "The athlete warms up with a lunge stretch, stepping forward with the left leg and raising both arms.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00376.npz", "sample_index": 376}
{"caption": "The athlete performs a warm-up movement where the right foot starts parallel to the left, then moves perpendicular and returns to parallel, repeating the cycle.", "dataset": "motionllama_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 146120, "index": 108, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00377.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/labels/fit3d/train/s04/smplx_slices/warmup_14_slice_2_retarget__seg0000__part0000.json", "segment_action": "The athlete performs a warm-up movement where the right foot starts parallel to the left, then moves perpendicular and returns to parallel, repeating the cycle.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "warmup_14_slice_2_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/g1/fit3d/train/s04/smplx_slices/warmup_14_slice_2_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The athlete performs a warm-up movement where the right foot starts parallel to the left, then moves perpendicular and returns to parallel, repeating the cycle.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00377.npz", "sample_index": 377}
{"caption": "He starts with his arms by his sides, ready to warm up, and then performs a series of coordinated movements, twisting to the left while raising his right arm overhead, before returning to an upright position and repeating the action.", "dataset": "motionllama_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 146167, "index": 155, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00378.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/labels/fit3d/train/s05/smplx_slices/warmup_4_slice_1_retarget__seg0000__part0001.json", "segment_action": "He starts with his arms by his sides, ready to warm up, and then performs a series of coordinated movements, twisting to the left while raising his right arm overhead, before returning to an upright position and repeating the action.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "warmup_4_slice_1_retarget__seg0000__part0001", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/g1/fit3d/train/s05/smplx_slices/warmup_4_slice_1_retarget__seg0000__part0001.npz", "split": "test", "text_path": null, "video_summary": "He starts with his arms by his sides, ready to warm up, and then performs a series of coordinated movements, twisting to the left while raising his right arm overhead, before returning to an upright position and repeating the action.", "window_end": 300, "window_start": 240}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00378.npz", "sample_index": 378}
{"caption": "The athlete stands at the box's bottom left corner with feet shoulder-width apart, lowering into a half-squat with a straight back and engaged core. As they hold this position, they rotate counterclockwise.", "dataset": "motionllama_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 146191, "index": 179, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00379.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/labels/fit3d/train/s07/smplx_slices/walk_the_box_slice_4_retarget__seg0000__part0001.json", "segment_action": "The athlete stands at the box's bottom left corner with feet shoulder-width apart, lowering into a half-squat with a straight back and engaged core. As they hold this position, they rotate counterclockwise.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "walk_the_box_slice_4_retarget__seg0000__part0001", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/g1/fit3d/train/s07/smplx_slices/walk_the_box_slice_4_retarget__seg0000__part0001.npz", "split": "test", "text_path": null, "video_summary": "The athlete stands at the box's bottom left corner with feet shoulder-width apart, lowering into a half-squat with a straight back and engaged core. As they hold this position, they rotate counterclockwise.", "window_end": 300, "window_start": 240}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00379.npz", "sample_index": 379}
{"caption": "He alternates arms while performing dumbbell curl trifectas.", "dataset": "motionllama_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 146223, "index": 211, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00380.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/labels/fit3d/train/s08/smplx_slices/dumbbell_curl_trifecta_slice_8_retarget__seg0000__part0000.json", "segment_action": "He alternates arms while performing dumbbell curl trifectas.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "dumbbell_curl_trifecta_slice_8_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/g1/fit3d/train/s08/smplx_slices/dumbbell_curl_trifecta_slice_8_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "He alternates arms while performing dumbbell curl trifectas.", "window_end": 180, "window_start": 120}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00380.npz", "sample_index": 380}
{"caption": "He alternates arms while performing dumbbell curl trifectas.", "dataset": "motionllama_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 146227, "index": 215, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00381.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/labels/fit3d/train/s08/smplx_slices/dumbbell_curl_trifecta_slice_8_retarget__seg0000__part0001.json", "segment_action": "He alternates arms while performing dumbbell curl trifectas.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "dumbbell_curl_trifecta_slice_8_retarget__seg0000__part0001", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/g1/fit3d/train/s08/smplx_slices/dumbbell_curl_trifecta_slice_8_retarget__seg0000__part0001.npz", "split": "test", "text_path": null, "video_summary": "He alternates arms while performing dumbbell curl trifectas.", "window_end": 300, "window_start": 240}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00381.npz", "sample_index": 381}
{"caption": "To initiate the exercise, stand on a firm surface with feet shoulder-width apart, maintaining an upright posture and engaging your core by tightening your abdominal muscles. Keep your arms relaxed at your sides with palms facing inward, ready for the movement. Next, lift your shoulders while squeezing your shoulder blades together to activate the shoulder muscles. As you continue, rotate your shoulders backward, forming a \"Y\" shape with your arms extended slightly backward and upward. Throughout the movement, maintain neck and head stability, avoiding tilting and focusing on the upper trapezius and deltoid muscles.", "dataset": "motionllama_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 146268, "index": 256, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00382.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/labels/fit3d/train/s09/smplx_slices/overhead_trap_raises_slice_1_retarget__seg0000__part0001.json", "segment_action": "To initiate the exercise, stand on a firm surface with feet shoulder-width apart, maintaining an upright posture and engaging your core by tightening your abdominal muscles. Keep your arms relaxed at your sides with palms facing inward, ready for the movement. Next, lift your shoulders while squeezing your shoulder blades together to activate the shoulder muscles. As you continue, rotate your shoulders backward, forming a \"Y\" shape with your arms extended slightly backward and upward. Throughout the movement, maintain neck and head stability, avoiding tilting and focusing on the upper trapezius and deltoid muscles.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "overhead_trap_raises_slice_1_retarget__seg0000__part0001", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/g1/fit3d/train/s09/smplx_slices/overhead_trap_raises_slice_1_retarget__seg0000__part0001.npz", "split": "test", "text_path": null, "video_summary": "To initiate the exercise, stand on a firm surface with feet shoulder-width apart, maintaining an upright posture and engaging your core by tightening your abdominal muscles. Keep your arms relaxed at your sides with palms facing inward, ready for the movement. Next, lift your shoulders while squeezing your shoulder blades together to activate the shoulder muscles. As you continue, rotate your shoulders backward, forming a \"Y\" shape with your arms extended slightly backward and upward. Throughout the movement, maintain neck and head stability, avoiding tilting and focusing on the upper trapezius and deltoid muscles.", "window_end": 180, "window_start": 120}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00382.npz", "sample_index": 382}
{"caption": "A person is doing overhead presses.", "dataset": "motionllama_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 146314, "index": 302, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00383.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/labels/fit3d/train/s10/smplx_slices/overhead_extension_thruster_slice_1_retarget__seg0000__part0000.json", "segment_action": "A person is doing overhead presses.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "overhead_extension_thruster_slice_1_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/g1/fit3d/train/s10/smplx_slices/overhead_extension_thruster_slice_1_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A person is doing overhead presses.", "window_end": 300, "window_start": 240}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00383.npz", "sample_index": 383}
{"caption": "The athlete performs high pulls in a half-squat stance, then lowers the barbell and stands up.", "dataset": "motionllama_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 146399, "index": 387, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00384.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/labels/fit3d/train/s11/smplx_slices/dumbbell_high_pulls_slice_2_retarget__seg0000__part0001.json", "segment_action": "The athlete performs high pulls in a half-squat stance, then lowers the barbell and stands up.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "dumbbell_high_pulls_slice_2_retarget__seg0000__part0001", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/g1/fit3d/train/s11/smplx_slices/dumbbell_high_pulls_slice_2_retarget__seg0000__part0001.npz", "split": "test", "text_path": null, "video_summary": "The athlete performs high pulls in a half-squat stance, then lowers the barbell and stands up.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00384.npz", "sample_index": 384}
{"caption": "An athlete started at the bottom left corner and walked the box five times in a counterclockwise direction.", "dataset": "motionllama_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 146410, "index": 398, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00385.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/labels/fit3d/train/s11/smplx_slices/walk_the_box_slice_1_retarget__seg0000__part0000.json", "segment_action": "An athlete started at the bottom left corner and walked the box five times in a counterclockwise direction.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "walk_the_box_slice_1_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/g1/fit3d/train/s11/smplx_slices/walk_the_box_slice_1_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "An athlete started at the bottom left corner and walked the box five times in a counterclockwise direction.", "window_end": 300, "window_start": 240}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00385.npz", "sample_index": 385}
{"caption": "A person sits, uses the right hand to open the laptop, types with both hands, and moves the phone with the right hand.", "dataset": "motionllama_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 146442, "index": 430, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00386.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/labels/trumans/slice_smplx_result/2023-01-14@22-22-31_slice_3_retarget__seg0000__part0000.json", "segment_action": "A person sits, uses the right hand to open the laptop, types with both hands, and moves the phone with the right hand.", "segment_frame_end": 291, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "2023-01-14@22-22-31_slice_3_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/g1/trumans/slice_smplx_result/2023-01-14@22-22-31_slice_3_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A person sits, uses the right hand to open the laptop, types with both hands, and moves the phone with the right hand.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00386.npz", "sample_index": 386}
{"caption": "A person sits down, then immediately gets up.", "dataset": "motionllama_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 146478, "index": 466, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00387.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/labels/trumans/slice_smplx_result/2023-01-14@23-52-23_slice_5_retarget__seg0000__part0000.json", "segment_action": "A person sits down, then immediately gets up.", "segment_frame_end": 210, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "2023-01-14@23-52-23_slice_5_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/g1/trumans/slice_smplx_result/2023-01-14@23-52-23_slice_5_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A person sits down, then immediately gets up.", "window_end": 135, "window_start": 75}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00387.npz", "sample_index": 387}
{"caption": "A person closes the microwave with their right hand, then places the pen and book down with their right hand and both hands respectively.", "dataset": "motionllama_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 146521, "index": 509, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00388.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/labels/trumans/slice_smplx_result/2023-01-15@22-23-46_slice_9_retarget__seg0000__part0000.json", "segment_action": "A person closes the microwave with their right hand, then places the pen and book down with their right hand and both hands respectively.", "segment_frame_end": 231, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "2023-01-15@22-23-46_slice_9_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/g1/trumans/slice_smplx_result/2023-01-15@22-23-46_slice_9_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A person closes the microwave with their right hand, then places the pen and book down with their right hand and both hands respectively.", "window_end": 146, "window_start": 86}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00388.npz", "sample_index": 388}
{"caption": "A person uses the left hand to both pick up and put down the phone.", "dataset": "motionllama_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 146530, "index": 518, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00389.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/labels/trumans/slice_smplx_result/2023-01-16@22-55-53_slice_4_retarget__seg0000__part0000.json", "segment_action": "A person uses the left hand to both pick up and put down the phone.", "segment_frame_end": 251, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "2023-01-16@22-55-53_slice_4_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/g1/trumans/slice_smplx_result/2023-01-16@22-55-53_slice_4_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A person uses the left hand to both pick up and put down the phone.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00389.npz", "sample_index": 389}
{"caption": "A person uses their right hand to pick up a cup, sits, drinks from it, and then gets up.", "dataset": "motionllama_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 146538, "index": 526, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00390.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/labels/trumans/slice_smplx_result/2023-01-16@23-01-39_slice_1_retarget__seg0000__part0000.json", "segment_action": "A person uses their right hand to pick up a cup, sits, drinks from it, and then gets up.", "segment_frame_end": 265, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "2023-01-16@23-01-39_slice_1_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/g1/trumans/slice_smplx_result/2023-01-16@23-01-39_slice_1_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A person uses their right hand to pick up a cup, sits, drinks from it, and then gets up.", "window_end": 265, "window_start": 205}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00390.npz", "sample_index": 390}
{"caption": "A person places the vase down with both hands before grasping the cup with their right hand.", "dataset": "motionllama_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 146558, "index": 546, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00391.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/labels/trumans/slice_smplx_result/2023-01-17@00-46-07_slice_3_retarget__seg0000__part0000.json", "segment_action": "A person places the vase down with both hands before grasping the cup with their right hand.", "segment_frame_end": 93, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "2023-01-17@00-46-07_slice_3_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/g1/trumans/slice_smplx_result/2023-01-17@00-46-07_slice_3_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A person places the vase down with both hands before grasping the cup with their right hand.", "window_end": 76, "window_start": 16}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00391.npz", "sample_index": 391}
{"caption": "A person takes a seat.", "dataset": "motionllama_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 146579, "index": 567, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00392.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/labels/trumans/slice_smplx_result/2023-01-17@01-35-33_slice_7_retarget__seg0000__part0000.json", "segment_action": "A person takes a seat.", "segment_frame_end": 90, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "2023-01-17@01-35-33_slice_7_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/g1/trumans/slice_smplx_result/2023-01-17@01-35-33_slice_7_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A person takes a seat.", "window_end": 75, "window_start": 15}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00392.npz", "sample_index": 392}
{"caption": "A person uses the right hand to put down the phone, pick up the handbag, sit down, put down the handbag, pick up the handbag, and put down the handbag.", "dataset": "motionllama_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 146605, "index": 593, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00393.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/labels/trumans/slice_smplx_result/2023-02-08@18-23-15_slice_8_retarget__seg0000__part0000.json", "segment_action": "A person uses the right hand to put down the phone, pick up the handbag, sit down, put down the handbag, pick up the handbag, and put down the handbag.", "segment_frame_end": 297, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "2023-02-08@18-23-15_slice_8_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/g1/trumans/slice_smplx_result/2023-02-08@18-23-15_slice_8_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A person uses the right hand to put down the phone, pick up the handbag, sit down, put down the handbag, pick up the handbag, and put down the handbag.", "window_end": 297, "window_start": 237}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00393.npz", "sample_index": 393}
{"caption": "A person uses their left hand to handle the phone and their right hand to handle the water bottle.", "dataset": "motionllama_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 146624, "index": 612, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00394.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/labels/trumans/slice_smplx_result/2023-02-08@21-18-24_slice_8_retarget__seg0000__part0000.json", "segment_action": "A person uses their left hand to handle the phone and their right hand to handle the water bottle.", "segment_frame_end": 276, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "2023-02-08@21-18-24_slice_8_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/g1/trumans/slice_smplx_result/2023-02-08@21-18-24_slice_8_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A person uses their left hand to handle the phone and their right hand to handle the water bottle.", "window_end": 276, "window_start": 216}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00394.npz", "sample_index": 394}
{"caption": "A person holds their phone with their right hand and drinks water with their left hand before placing the phone down.", "dataset": "motionllama_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 146641, "index": 629, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00395.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/labels/trumans/slice_smplx_result/2023-02-09@14-04-02_slice_9_retarget__seg0000__part0000.json", "segment_action": "A person holds their phone with their right hand and drinks water with their left hand before placing the phone down.", "segment_frame_end": 183, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "2023-02-09@14-04-02_slice_9_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/g1/trumans/slice_smplx_result/2023-02-09@14-04-02_slice_9_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A person holds their phone with their right hand and drinks water with their left hand before placing the phone down.", "window_end": 122, "window_start": 62}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00395.npz", "sample_index": 395}
{"caption": "The person closes the book with both hands before picking up the pen with the right hand.", "dataset": "motionllama_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 146706, "index": 694, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00396.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/labels/trumans/slice_smplx_result/2023-02-12@12-07-11_slice_10_retarget__seg0000__part0000.json", "segment_action": "The person closes the book with both hands before picking up the pen with the right hand.", "segment_frame_end": 243, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "2023-02-12@12-07-11_slice_10_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/g1/trumans/slice_smplx_result/2023-02-12@12-07-11_slice_10_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The person closes the book with both hands before picking up the pen with the right hand.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00396.npz", "sample_index": 396}
{"caption": "The person grasps the water bottle with the right hand, gets up, and sets it down with the left hand.", "dataset": "motionllama_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 146766, "index": 754, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00397.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/labels/trumans/slice_smplx_result/2023-02-12@15-29-58_slice_7_retarget__seg0000__part0000.json", "segment_action": "The person grasps the water bottle with the right hand, gets up, and sets it down with the left hand.", "segment_frame_end": 260, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "2023-02-12@15-29-58_slice_7_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/g1/trumans/slice_smplx_result/2023-02-12@15-29-58_slice_7_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The person grasps the water bottle with the right hand, gets up, and sets it down with the left hand.", "window_end": 260, "window_start": 200}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00397.npz", "sample_index": 397}
{"caption": "A person writes with the right hand and then uses the right hand to put down the pen.", "dataset": "motionllama_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 146786, "index": 774, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00398.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/labels/trumans/slice_smplx_result/2023-02-12@16-11-02_slice_4_retarget__seg0000__part0000.json", "segment_action": "A person writes with the right hand and then uses the right hand to put down the pen.", "segment_frame_end": 258, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "2023-02-12@16-11-02_slice_4_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/g1/trumans/slice_smplx_result/2023-02-12@16-11-02_slice_4_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A person writes with the right hand and then uses the right hand to put down the pen.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00398.npz", "sample_index": 398}
{"caption": "A person stands up, takes the pen with the right hand, and then sits down.", "dataset": "motionllama_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 146867, "index": 855, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00399.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/labels/trumans/slice_smplx_result/2023-02-12@22-16-01_slice_7_retarget__seg0000__part0000.json", "segment_action": "A person stands up, takes the pen with the right hand, and then sits down.", "segment_frame_end": 210, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "2023-02-12@22-16-01_slice_7_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/g1/trumans/slice_smplx_result/2023-02-12@22-16-01_slice_7_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A person stands up, takes the pen with the right hand, and then sits down.", "window_end": 135, "window_start": 75}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00399.npz", "sample_index": 399}
{"caption": "A person lifts and moves a chair with both hands, and then handles and sets down a keyboard with both hands.", "dataset": "motionllama_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 146874, "index": 862, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00400.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/labels/trumans/slice_smplx_result/2023-02-13@14-47-52_slice_6_retarget__seg0000__part0000.json", "segment_action": "A person lifts and moves a chair with both hands, and then handles and sets down a keyboard with both hands.", "segment_frame_end": 181, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "2023-02-13@14-47-52_slice_6_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/g1/trumans/slice_smplx_result/2023-02-13@14-47-52_slice_6_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A person lifts and moves a chair with both hands, and then handles and sets down a keyboard with both hands.", "window_end": 120, "window_start": 60}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00400.npz", "sample_index": 400}
{"caption": "A person is pushing a chair with both hands.", "dataset": "motionllama_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 146941, "index": 929, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00401.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/labels/trumans/slice_smplx_result/2023-02-13@17-52-39_slice_5_retarget__seg0000__part0000.json", "segment_action": "A person is pushing a chair with both hands.", "segment_frame_end": 187, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "2023-02-13@17-52-39_slice_5_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/g1/trumans/slice_smplx_result/2023-02-13@17-52-39_slice_5_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A person is pushing a chair with both hands.", "window_end": 124, "window_start": 64}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00401.npz", "sample_index": 401}
{"caption": "The person uses the left hand to pick up the keyboard, then squats down, and continues to use it.", "dataset": "motionllama_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 146960, "index": 948, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00402.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/labels/trumans/slice_smplx_result/2023-02-13@21-27-28_slice_8_retarget__seg0000__part0000.json", "segment_action": "The person uses the left hand to pick up the keyboard, then squats down, and continues to use it.", "segment_frame_end": 270, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "2023-02-13@21-27-28_slice_8_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/g1/trumans/slice_smplx_result/2023-02-13@21-27-28_slice_8_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The person uses the left hand to pick up the keyboard, then squats down, and continues to use it.", "window_end": 270, "window_start": 210}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00402.npz", "sample_index": 402}
{"caption": "A person uses their left hand to pick up the mouse and phone, then puts down the mouse and the phone with their left and right hands respectively.", "dataset": "motionllama_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 146967, "index": 955, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00403.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/labels/trumans/slice_smplx_result/2023-02-13@22-04-21_slice_1_retarget__seg0000__part0000.json", "segment_action": "A person uses their left hand to pick up the mouse and phone, then puts down the mouse and the phone with their left and right hands respectively.", "segment_frame_end": 282, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "2023-02-13@22-04-21_slice_1_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/g1/trumans/slice_smplx_result/2023-02-13@22-04-21_slice_1_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A person uses their left hand to pick up the mouse and phone, then puts down the mouse and the phone with their left and right hands respectively.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00403.npz", "sample_index": 403}
{"caption": "A person sits down, placing their right hand on the table and using both hands to type on the keyboard.", "dataset": "motionllama_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 146981, "index": 969, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00404.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/labels/trumans/slice_smplx_result/2023-02-13@22-32-46_slice_4_retarget__seg0000__part0000.json", "segment_action": "A person sits down, placing their right hand on the table and using both hands to type on the keyboard.", "segment_frame_end": 231, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "2023-02-13@22-32-46_slice_4_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/g1/trumans/slice_smplx_result/2023-02-13@22-32-46_slice_4_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A person sits down, placing their right hand on the table and using both hands to type on the keyboard.", "window_end": 231, "window_start": 171}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00404.npz", "sample_index": 404}
{"caption": "A person operates the oven with one hand and holds a cup with the other, before taking a seat.", "dataset": "motionllama_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 146983, "index": 971, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00405.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/labels/trumans/slice_smplx_result/2023-02-13@22-51-56_slice_4_retarget__seg0000__part0000.json", "segment_action": "A person operates the oven with one hand and holds a cup with the other, before taking a seat.", "segment_frame_end": 207, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "2023-02-13@22-51-56_slice_4_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/g1/trumans/slice_smplx_result/2023-02-13@22-51-56_slice_4_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A person operates the oven with one hand and holds a cup with the other, before taking a seat.", "window_end": 134, "window_start": 74}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00405.npz", "sample_index": 405}
{"caption": "The person closes the laptop with their right hand, then uses their left hand to pick up and put down the cup.", "dataset": "motionllama_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 147008, "index": 996, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00406.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/labels/trumans/slice_smplx_result/2023-02-14@12-10-47_slice_5_retarget__seg0000__part0000.json", "segment_action": "The person closes the laptop with their right hand, then uses their left hand to pick up and put down the cup.", "segment_frame_end": 264, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "2023-02-14@12-10-47_slice_5_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/g1/trumans/slice_smplx_result/2023-02-14@12-10-47_slice_5_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The person closes the laptop with their right hand, then uses their left hand to pick up and put down the cup.", "window_end": 264, "window_start": 204}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00406.npz", "sample_index": 406}
{"caption": "A person uses their left hand to pick up, drink from, and then set down a water bottle after standing up.", "dataset": "motionllama_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 147009, "index": 997, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00407.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/labels/trumans/slice_smplx_result/2023-02-14@12-16-02_slice_3_retarget__seg0000__part0000.json", "segment_action": "A person uses their left hand to pick up, drink from, and then set down a water bottle after standing up.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "2023-02-14@12-16-02_slice_3_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/g1/trumans/slice_smplx_result/2023-02-14@12-16-02_slice_3_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A person uses their left hand to pick up, drink from, and then set down a water bottle after standing up.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00407.npz", "sample_index": 407}
{"caption": "A person uses their right hand to pick up the water bottle, switches to their left hand to open the drawer, then closes it with the same hand, reopens it, and switches back to their right hand to pick up the water bottle.", "dataset": "motionllama_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 147049, "index": 1037, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00408.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/labels/trumans/slice_smplx_result/2023-02-16@15-17-47_slice_6_retarget__seg0000__part0000.json", "segment_action": "A person uses their right hand to pick up the water bottle, switches to their left hand to open the drawer, then closes it with the same hand, reopens it, and switches back to their right hand to pick up the water bottle.", "segment_frame_end": 286, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "2023-02-16@15-17-47_slice_6_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/g1/trumans/slice_smplx_result/2023-02-16@15-17-47_slice_6_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A person uses their right hand to pick up the water bottle, switches to their left hand to open the drawer, then closes it with the same hand, reopens it, and switches back to their right hand to pick up the water bottle.", "window_end": 286, "window_start": 226}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00408.npz", "sample_index": 408}
{"caption": "A person sits, places the phone with the right hand, then grasps the water bottle with the right hand.", "dataset": "motionllama_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 147073, "index": 1061, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00409.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/labels/trumans/slice_smplx_result/2023-02-16@15-56-36_slice_2_retarget__seg0000__part0000.json", "segment_action": "A person sits, places the phone with the right hand, then grasps the water bottle with the right hand.", "segment_frame_end": 276, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "2023-02-16@15-56-36_slice_2_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/g1/trumans/slice_smplx_result/2023-02-16@15-56-36_slice_2_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A person sits, places the phone with the right hand, then grasps the water bottle with the right hand.", "window_end": 276, "window_start": 216}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00409.npz", "sample_index": 409}
{"caption": "A person uses the left hand for the handbag, right hand for the mouse, and left hand for water.", "dataset": "motionllama_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 147127, "index": 1115, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00410.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/labels/trumans/slice_smplx_result/2023-02-16@17-41-45_slice_3_retarget__seg0000__part0000.json", "segment_action": "A person uses the left hand for the handbag, right hand for the mouse, and left hand for water.", "segment_frame_end": 294, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "2023-02-16@17-41-45_slice_3_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/g1/trumans/slice_smplx_result/2023-02-16@17-41-45_slice_3_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A person uses the left hand for the handbag, right hand for the mouse, and left hand for water.", "window_end": 177, "window_start": 117}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00410.npz", "sample_index": 410}
{"caption": "The person uses their right hand to pick up and put down the keyboard, open, and close the drawer.", "dataset": "motionllama_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 147130, "index": 1118, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00411.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/labels/trumans/slice_smplx_result/2023-02-16@17-47-21_slice_4_retarget__seg0000__part0000.json", "segment_action": "The person uses their right hand to pick up and put down the keyboard, open, and close the drawer.", "segment_frame_end": 255, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "2023-02-16@17-47-21_slice_4_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/g1/trumans/slice_smplx_result/2023-02-16@17-47-21_slice_4_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The person uses their right hand to pick up and put down the keyboard, open, and close the drawer.", "window_end": 158, "window_start": 98}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00411.npz", "sample_index": 411}
{"caption": "A person uses their right hand to pick up the briefcase, their left hand to put down the keyboard, their right hand to put down the briefcase, and their right hand to open the drawer.", "dataset": "motionllama_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 147134, "index": 1122, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00412.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/labels/trumans/slice_smplx_result/2023-02-16@17-50-11_slice_6_retarget__seg0000__part0000.json", "segment_action": "A person uses their right hand to pick up the briefcase, their left hand to put down the keyboard, their right hand to put down the briefcase, and their right hand to open the drawer.", "segment_frame_end": 284, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "2023-02-16@17-50-11_slice_6_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/g1/trumans/slice_smplx_result/2023-02-16@17-50-11_slice_6_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A person uses their right hand to pick up the briefcase, their left hand to put down the keyboard, their right hand to put down the briefcase, and their right hand to open the drawer.", "window_end": 284, "window_start": 224}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00412.npz", "sample_index": 412}
{"caption": "A person places the cup down with their right hand and then sets their right hand on the stool.", "dataset": "motionllama_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 147138, "index": 1126, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00413.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/labels/trumans/slice_smplx_result/2023-02-17@09-58-24_slice_7_retarget__seg0000__part0000.json", "segment_action": "A person places the cup down with their right hand and then sets their right hand on the stool.", "segment_frame_end": 126, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "2023-02-17@09-58-24_slice_7_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/g1/trumans/slice_smplx_result/2023-02-17@09-58-24_slice_7_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A person places the cup down with their right hand and then sets their right hand on the stool.", "window_end": 126, "window_start": 66}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00413.npz", "sample_index": 413}
{"caption": "A person drinks water with his right hand.", "dataset": "motionllama_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 147159, "index": 1147, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00414.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/labels/trumans/slice_smplx_result/2023-02-17@10-28-43_slice_8_retarget__seg0000__part0000.json", "segment_action": "A person drinks water with his right hand.", "segment_frame_end": 171, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "2023-02-17@10-28-43_slice_8_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/g1/trumans/slice_smplx_result/2023-02-17@10-28-43_slice_8_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A person drinks water with his right hand.", "window_end": 171, "window_start": 111}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00414.npz", "sample_index": 414}
{"caption": "A person uses their left hand to open and close the cabinet.", "dataset": "motionllama_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 147169, "index": 1157, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00415.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/labels/trumans/slice_smplx_result/2023-02-17@10-52-40_slice_4_retarget__seg0000__part0000.json", "segment_action": "A person uses their left hand to open and close the cabinet.", "segment_frame_end": 127, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "2023-02-17@10-52-40_slice_4_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/g1/trumans/slice_smplx_result/2023-02-17@10-52-40_slice_4_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A person uses their left hand to open and close the cabinet.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00415.npz", "sample_index": 415}
{"caption": "He places the cup down with his right hand before grasping the vase with both hands.", "dataset": "motionllama_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 147186, "index": 1174, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00416.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/labels/trumans/slice_smplx_result/2023-02-17@11-05-08_slice_2_retarget__seg0000__part0000.json", "segment_action": "He places the cup down with his right hand before grasping the vase with both hands.", "segment_frame_end": 164, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "2023-02-17@11-05-08_slice_2_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/g1/trumans/slice_smplx_result/2023-02-17@11-05-08_slice_2_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "He places the cup down with his right hand before grasping the vase with both hands.", "window_end": 164, "window_start": 104}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00416.npz", "sample_index": 416}
{"caption": "A person pours water and puts down the cup with their right hand before sitting down.", "dataset": "motionllama_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 147207, "index": 1195, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00417.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/labels/trumans/slice_smplx_result/2023-02-17@11-55-56_slice_4_retarget__seg0000__part0000.json", "segment_action": "A person pours water and puts down the cup with their right hand before sitting down.", "segment_frame_end": 281, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "2023-02-17@11-55-56_slice_4_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/g1/trumans/slice_smplx_result/2023-02-17@11-55-56_slice_4_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A person pours water and puts down the cup with their right hand before sitting down.", "window_end": 281, "window_start": 221}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00417.npz", "sample_index": 417}
{"caption": "A person grasps the keyboard, sets it down, operates the microwave with one hand, and lifts a pen with the other hand.", "dataset": "motionllama_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 147292, "index": 1280, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00418.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/labels/trumans/slice_smplx_result/2023-02-18@21-15-05_slice_4_retarget__seg0000__part0000.json", "segment_action": "A person grasps the keyboard, sets it down, operates the microwave with one hand, and lifts a pen with the other hand.", "segment_frame_end": 270, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "2023-02-18@21-15-05_slice_4_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/g1/trumans/slice_smplx_result/2023-02-18@21-15-05_slice_4_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A person grasps the keyboard, sets it down, operates the microwave with one hand, and lifts a pen with the other hand.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00418.npz", "sample_index": 418}
{"caption": "He drinks water with his right hand, sets the bottle down, and stands up.", "dataset": "motionllama_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 147371, "index": 1359, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00419.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/labels/trumans/slice_smplx_result/2023-02-19@22-43-27_slice_4_retarget__seg0000__part0000.json", "segment_action": "He drinks water with his right hand, sets the bottle down, and stands up.", "segment_frame_end": 177, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "2023-02-19@22-43-27_slice_4_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/motionllama/g1/trumans/slice_smplx_result/2023-02-19@22-43-27_slice_4_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "He drinks water with his right hand, sets the bottle down, and stands up.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00419.npz", "sample_index": 419}
{"caption": "sks person performs a series of throws with an object.", "dataset": "permo_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 147438, "index": 49, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00420.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/labels/Angry_Throw_A02_003_retarget__seg0000__part0000.json", "segment_action": "sks person performs a series of throws with an object.", "segment_frame_end": 144, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "Angry_Throw_A02_003_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/g1/Angry_Throw_A02_003_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "sks person performs a series of throws with an object.", "window_end": 144, "window_start": 84}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00420.npz", "sample_index": 420}
{"caption": "sks person runs straight forward.", "dataset": "permo_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 147558, "index": 169, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00421.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/labels/Ballerina_Run_A02_004_retarget__seg0000__part0000.json", "segment_action": "sks person runs straight forward.", "segment_frame_end": 62, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "Ballerina_Run_A02_004_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/g1/Ballerina_Run_A02_004_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "sks person runs straight forward.", "window_end": 62, "window_start": 2}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00421.npz", "sample_index": 421}
{"caption": "sks hops forward several times on one leg.", "dataset": "permo_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 147584, "index": 195, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00422.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/labels/Childish_Hop_A04_001_retarget__seg0000__part0000.json", "segment_action": "sks hops forward several times on one leg.", "segment_frame_end": 83, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "Childish_Hop_A04_001_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/g1/Childish_Hop_A04_001_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "sks hops forward several times on one leg.", "window_end": 72, "window_start": 12}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00422.npz", "sample_index": 422}
{"caption": "sks hops forward several times on one leg.", "dataset": "permo_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 147585, "index": 196, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00423.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/labels/Childish_Hop_A04_001_retarget__seg0000__part0000.json", "segment_action": "sks hops forward several times on one leg.", "segment_frame_end": 83, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "Childish_Hop_A04_001_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/g1/Childish_Hop_A04_001_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "sks hops forward several times on one leg.", "window_end": 83, "window_start": 23}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00423.npz", "sample_index": 423}
{"caption": "sks person is jumping ahead", "dataset": "permo_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 147638, "index": 249, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00424.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/labels/Cold_Hop_A02_004_retarget__seg0000__part0000.json", "segment_action": "sks person is jumping ahead", "segment_frame_end": 69, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "Cold_Hop_A02_004_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/g1/Cold_Hop_A02_004_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "sks person is jumping ahead", "window_end": 69, "window_start": 9}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00424.npz", "sample_index": 424}
{"caption": "sks person moves forward through a series of jumps.", "dataset": "permo_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 147689, "index": 300, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00425.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/labels/Crowded_Jump_A01_002_retarget__seg0000__part0000.json", "segment_action": "sks person moves forward through a series of jumps.", "segment_frame_end": 104, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "Crowded_Jump_A01_002_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/g1/Crowded_Jump_A01_002_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "sks person moves forward through a series of jumps.", "window_end": 104, "window_start": 44}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00425.npz", "sample_index": 425}
{"caption": "sks person performs forward kicks.", "dataset": "permo_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 147700, "index": 311, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00426.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/labels/Crowded_Kick_A03_002_retarget__seg0000__part0000.json", "segment_action": "sks person performs forward kicks.", "segment_frame_end": 149, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "Crowded_Kick_A03_002_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/g1/Crowded_Kick_A03_002_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "sks person performs forward kicks.", "window_end": 104, "window_start": 44}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00426.npz", "sample_index": 426}
{"caption": "sks person walks ahead and then shifts direction sideways.", "dataset": "permo_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 147721, "index": 332, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00427.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/labels/Crowded_Transition_A03_002_retarget__seg0000__part0000.json", "segment_action": "sks person walks ahead and then shifts direction sideways.", "segment_frame_end": 85, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "Crowded_Transition_A03_002_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/g1/Crowded_Transition_A03_002_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "sks person walks ahead and then shifts direction sideways.", "window_end": 72, "window_start": 12}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00427.npz", "sample_index": 427}
{"caption": "sks person targets an object on the floor with several kicks.", "dataset": "permo_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 147769, "index": 380, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00428.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/labels/Drunken_KickSth_A04_004_retarget__seg0000__part0000.json", "segment_action": "sks person targets an object on the floor with several kicks.", "segment_frame_end": 194, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "Drunken_KickSth_A04_004_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/g1/Drunken_KickSth_A04_004_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "sks person targets an object on the floor with several kicks.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00428.npz", "sample_index": 428}
{"caption": "Throwing several times, sks person targets a point.", "dataset": "permo_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 147788, "index": 399, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00429.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/labels/Drunken_Throw_A04_001_retarget__seg0000__part0000.json", "segment_action": "Throwing several times, sks person targets a point.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "Drunken_Throw_A04_001_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/g1/Drunken_Throw_A04_001_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "Throwing several times, sks person targets a point.", "window_end": 180, "window_start": 120}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00429.npz", "sample_index": 429}
{"caption": "sks person is walking forward.", "dataset": "permo_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 147861, "index": 472, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00430.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/labels/Elegant_Walk_A03_002_retarget__seg0000__part0000.json", "segment_action": "sks person is walking forward.", "segment_frame_end": 117, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "Elegant_Walk_A03_002_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/g1/Elegant_Walk_A03_002_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "sks person is walking forward.", "window_end": 117, "window_start": 57}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00430.npz", "sample_index": 430}
{"caption": "sks person waves with one arm repeatedly.", "dataset": "permo_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 147873, "index": 484, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00431.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/labels/Elegant_Wave_A03_002_retarget__seg0000__part0000.json", "segment_action": "sks person waves with one arm repeatedly.", "segment_frame_end": 132, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "Elegant_Wave_A03_002_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/g1/Elegant_Wave_A03_002_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "sks person waves with one arm repeatedly.", "window_end": 132, "window_start": 72}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00431.npz", "sample_index": 431}
{"caption": "sks person performs several throws with an item.", "dataset": "permo_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 147901, "index": 512, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00432.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/labels/Exhausted_Throw_A04_004_retarget__seg0000__part0000.json", "segment_action": "sks person performs several throws with an item.", "segment_frame_end": 251, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "Exhausted_Throw_A04_004_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/g1/Exhausted_Throw_A04_004_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "sks person performs several throws with an item.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00432.npz", "sample_index": 432}
{"caption": "sks walks forward and then makes a sideways turn.", "dataset": "permo_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 147910, "index": 521, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00433.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/labels/Exhausted_Transition_A05_002_retarget__seg0000__part0000.json", "segment_action": "sks walks forward and then makes a sideways turn.", "segment_frame_end": 122, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "Exhausted_Transition_A05_002_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/g1/Exhausted_Transition_A05_002_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "sks walks forward and then makes a sideways turn.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00433.npz", "sample_index": 433}
{"caption": "sks person hops several times, moving on one leg.", "dataset": "permo_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 147941, "index": 552, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00434.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/labels/Fearful_Hop_A05_001_retarget__seg0000__part0000.json", "segment_action": "sks person hops several times, moving on one leg.", "segment_frame_end": 98, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "Fearful_Hop_A05_001_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/g1/Fearful_Hop_A05_001_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "sks person hops several times, moving on one leg.", "window_end": 79, "window_start": 19}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00434.npz", "sample_index": 434}
{"caption": "sks person hits something on the ground with several kicks.", "dataset": "permo_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 147952, "index": 563, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00435.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/labels/Fearful_KickSth_A03_004_retarget__seg0000__part0000.json", "segment_action": "sks person hits something on the ground with several kicks.", "segment_frame_end": 143, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "Fearful_KickSth_A03_004_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/g1/Fearful_KickSth_A03_004_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "sks person hits something on the ground with several kicks.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00435.npz", "sample_index": 435}
{"caption": "sks person walks forward and then turns to the side.", "dataset": "permo_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 147983, "index": 594, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00436.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/labels/Fearful_Transition_A04_003_retarget__seg0000__part0000.json", "segment_action": "sks person walks forward and then turns to the side.", "segment_frame_end": 106, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "Fearful_Transition_A04_003_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/g1/Fearful_Transition_A04_003_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "sks person walks forward and then turns to the side.", "window_end": 106, "window_start": 46}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00436.npz", "sample_index": 436}
{"caption": "sks person leaps, using only one leg to move.", "dataset": "permo_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 148012, "index": 623, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00437.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/labels/Happy_Hop_A02_002_retarget__seg0000__part0000.json", "segment_action": "sks person leaps, using only one leg to move.", "segment_frame_end": 73, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "Happy_Hop_A02_002_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/g1/Happy_Hop_A02_002_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "sks person leaps, using only one leg to move.", "window_end": 66, "window_start": 6}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00437.npz", "sample_index": 437}
{"caption": "sks person hops forward repeatedly, staying balanced on one leg.", "dataset": "permo_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 148015, "index": 626, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00438.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/labels/Happy_Hop_A05_002_retarget__seg0000__part0000.json", "segment_action": "sks person hops forward repeatedly, staying balanced on one leg.", "segment_frame_end": 78, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "Happy_Hop_A05_002_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/g1/Happy_Hop_A05_002_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "sks person hops forward repeatedly, staying balanced on one leg.", "window_end": 69, "window_start": 9}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00438.npz", "sample_index": 438}
{"caption": "sks person propels themselves forward with several hops.", "dataset": "permo_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 148124, "index": 735, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00439.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/labels/Healthy_Jump_A04_004_retarget__seg0000__part0000.json", "segment_action": "sks person propels themselves forward with several hops.", "segment_frame_end": 180, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "Healthy_Jump_A04_004_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/g1/Healthy_Jump_A04_004_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "sks person propels themselves forward with several hops.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00439.npz", "sample_index": 439}
{"caption": "sks person repeatedly waves with one arm in a greeting gesture.", "dataset": "permo_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 148211, "index": 822, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00440.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/labels/Hulk_Wave_A03_001_retarget__seg0000__part0000.json", "segment_action": "sks person repeatedly waves with one arm in a greeting gesture.", "segment_frame_end": 131, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "Hulk_Wave_A03_001_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/g1/Hulk_Wave_A03_001_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "sks person repeatedly waves with one arm in a greeting gesture.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00440.npz", "sample_index": 440}
{"caption": "sks person hops ahead.", "dataset": "permo_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 148215, "index": 826, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00441.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/labels/Legaching_Hop_A01_003_retarget__seg0000__part0000.json", "segment_action": "sks person hops ahead.", "segment_frame_end": 98, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "Legaching_Hop_A01_003_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/g1/Legaching_Hop_A01_003_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "sks person hops ahead.", "window_end": 79, "window_start": 19}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00441.npz", "sample_index": 441}
{"caption": "sks repeatedly throws an object towards a target.", "dataset": "permo_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 148279, "index": 890, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00442.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/labels/Monkey_Throw_A02_002_retarget__seg0000__part0000.json", "segment_action": "sks repeatedly throws an object towards a target.", "segment_frame_end": 159, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "Monkey_Throw_A02_002_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/g1/Monkey_Throw_A02_002_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "sks repeatedly throws an object towards a target.", "window_end": 110, "window_start": 50}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00442.npz", "sample_index": 442}
{"caption": "Hopping on a single foot, sks person moves ahead.", "dataset": "permo_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 148319, "index": 930, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00443.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/labels/Neutral_Hop_A05_001_retarget__seg0000__part0000.json", "segment_action": "Hopping on a single foot, sks person moves ahead.", "segment_frame_end": 82, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "Neutral_Hop_A05_001_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/g1/Neutral_Hop_A05_001_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "Hopping on a single foot, sks person moves ahead.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00443.npz", "sample_index": 443}
{"caption": "Hopping on a single foot, sks person moves ahead.", "dataset": "permo_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 148320, "index": 931, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00444.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/labels/Neutral_Hop_A05_001_retarget__seg0000__part0000.json", "segment_action": "Hopping on a single foot, sks person moves ahead.", "segment_frame_end": 82, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "Neutral_Hop_A05_001_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/g1/Neutral_Hop_A05_001_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "Hopping on a single foot, sks person moves ahead.", "window_end": 71, "window_start": 11}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00444.npz", "sample_index": 444}
{"caption": "sks person kicks the floor.", "dataset": "permo_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 148337, "index": 948, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00445.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/labels/Neutral_KickSth_A02_005_retarget__seg0000__part0000.json", "segment_action": "sks person kicks the floor.", "segment_frame_end": 187, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "Neutral_KickSth_A02_005_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/g1/Neutral_KickSth_A02_005_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "sks person kicks the floor.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00445.npz", "sample_index": 445}
{"caption": "sks person walks directly ahead without changing direction.", "dataset": "permo_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 148408, "index": 1019, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00446.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/labels/Ninja_Walk_A04_004_retarget__seg0000__part0000.json", "segment_action": "sks person walks directly ahead without changing direction.", "segment_frame_end": 146, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "Ninja_Walk_A04_004_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/g1/Ninja_Walk_A04_004_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "sks person walks directly ahead without changing direction.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00446.npz", "sample_index": 446}
{"caption": "sks person moves ahead with a series of jumps.", "dataset": "permo_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 148430, "index": 1041, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00447.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/labels/Old_Jump_A05_002_retarget__seg0000__part0000.json", "segment_action": "sks person moves ahead with a series of jumps.", "segment_frame_end": 281, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "Old_Jump_A05_002_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/g1/Old_Jump_A05_002_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "sks person moves ahead with a series of jumps.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00447.npz", "sample_index": 447}
{"caption": "sks person kicks several times, sending their foot into empty space.", "dataset": "permo_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 148434, "index": 1045, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00448.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/labels/Old_Kick_A01_002_retarget__seg0000__part0000.json", "segment_action": "sks person kicks several times, sending their foot into empty space.", "segment_frame_end": 202, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "Old_Kick_A01_002_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/g1/Old_Kick_A01_002_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "sks person kicks several times, sending their foot into empty space.", "window_end": 131, "window_start": 71}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00448.npz", "sample_index": 448}
{"caption": "sks person repeatedly kicks forward with their foot.", "dataset": "permo_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 148482, "index": 1093, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00449.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/labels/Penguin_Kick_A03_002_retarget__seg0000__part0000.json", "segment_action": "sks person repeatedly kicks forward with their foot.", "segment_frame_end": 133, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "Penguin_Kick_A03_002_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/g1/Penguin_Kick_A03_002_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "sks person repeatedly kicks forward with their foot.", "window_end": 96, "window_start": 36}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00449.npz", "sample_index": 449}
{"caption": "sks person makes several forward punches.", "dataset": "permo_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 148486, "index": 1097, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00450.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/labels/Penguin_Punch_A04_003_retarget__seg0000__part0000.json", "segment_action": "sks person makes several forward punches.", "segment_frame_end": 164, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "Penguin_Punch_A04_003_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/g1/Penguin_Punch_A04_003_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "sks person makes several forward punches.", "window_end": 164, "window_start": 104}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00450.npz", "sample_index": 450}
{"caption": "sks walks forward, then executes a sideways turn.", "dataset": "permo_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 148501, "index": 1112, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00451.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/labels/Penguin_Transition_A04_002_retarget__seg0000__part0000.json", "segment_action": "sks walks forward, then executes a sideways turn.", "segment_frame_end": 122, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "Penguin_Transition_A04_002_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/g1/Penguin_Transition_A04_002_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "sks walks forward, then executes a sideways turn.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00451.npz", "sample_index": 451}
{"caption": "sks walks forward, then executes a sideways turn.", "dataset": "permo_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 148503, "index": 1114, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00452.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/labels/Penguin_Transition_A04_002_retarget__seg0000__part0000.json", "segment_action": "sks walks forward, then executes a sideways turn.", "segment_frame_end": 122, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "Penguin_Transition_A04_002_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/g1/Penguin_Transition_A04_002_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "sks walks forward, then executes a sideways turn.", "window_end": 122, "window_start": 62}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00452.npz", "sample_index": 452}
{"caption": "Waving several times, sks person uses one arm to greet.", "dataset": "permo_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 148584, "index": 1195, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00453.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/labels/Robot_Wave_A02_001_retarget__seg0000__part0000.json", "segment_action": "Waving several times, sks person uses one arm to greet.", "segment_frame_end": 104, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "Robot_Wave_A02_001_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/g1/Robot_Wave_A02_001_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "Waving several times, sks person uses one arm to greet.", "window_end": 104, "window_start": 44}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00453.npz", "sample_index": 453}
{"caption": "sks walks ahead and then transitions sideways.", "dataset": "permo_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 148634, "index": 1245, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00454.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/labels/SWAT_Transition_A02_001_retarget__seg0000__part0000.json", "segment_action": "sks walks ahead and then transitions sideways.", "segment_frame_end": 127, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "SWAT_Transition_A02_001_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/g1/SWAT_Transition_A02_001_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "sks walks ahead and then transitions sideways.", "window_end": 94, "window_start": 34}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00454.npz", "sample_index": 454}
{"caption": "sks person leaps, landing on the same leg each time.", "dataset": "permo_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 148706, "index": 1317, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00455.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/labels/Shy_Hop_A01_004_retarget__seg0000__part0000.json", "segment_action": "sks person leaps, landing on the same leg each time.", "segment_frame_end": 76, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "Shy_Hop_A01_004_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/g1/Shy_Hop_A01_004_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "sks person leaps, landing on the same leg each time.", "window_end": 76, "window_start": 16}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00455.npz", "sample_index": 455}
{"caption": "sks targets an object on the floor with a series of kicks.", "dataset": "permo_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 148722, "index": 1333, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00456.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/labels/Shy_KickSth_A01_004_retarget__seg0000__part0000.json", "segment_action": "sks targets an object on the floor with a series of kicks.", "segment_frame_end": 183, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "Shy_KickSth_A01_004_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/g1/Shy_KickSth_A01_004_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "sks targets an object on the floor with a series of kicks.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00456.npz", "sample_index": 456}
{"caption": "Throwing multiple times, sks person propels an object through the air.", "dataset": "permo_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 148746, "index": 1357, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00457.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/labels/Shy_Throw_A02_003_retarget__seg0000__part0000.json", "segment_action": "Throwing multiple times, sks person propels an object through the air.", "segment_frame_end": 195, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "Shy_Throw_A02_003_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/g1/Shy_Throw_A02_003_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "Throwing multiple times, sks person propels an object through the air.", "window_end": 128, "window_start": 68}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00457.npz", "sample_index": 457}
{"caption": "sks person repeatedly springs forward.", "dataset": "permo_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 148760, "index": 1371, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00458.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/labels/Silly_Jump_A01_002_retarget__seg0000__part0000.json", "segment_action": "sks person repeatedly springs forward.", "segment_frame_end": 78, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "Silly_Jump_A01_002_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/g1/Silly_Jump_A01_002_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "sks person repeatedly springs forward.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00458.npz", "sample_index": 458}
{"caption": "sks person moves ahead with a series of jumps.", "dataset": "permo_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 148768, "index": 1379, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00459.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/labels/Silly_Jump_A04_004_retarget__seg0000__part0000.json", "segment_action": "sks person moves ahead with a series of jumps.", "segment_frame_end": 104, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "Silly_Jump_A04_004_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/g1/Silly_Jump_A04_004_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "sks person moves ahead with a series of jumps.", "window_end": 104, "window_start": 44}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00459.npz", "sample_index": 459}
{"caption": "sks person hops several times, using one leg to move forward.", "dataset": "permo_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 148807, "index": 1418, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00460.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/labels/Strained_Hop_A03_001_retarget__seg0000__part0000.json", "segment_action": "sks person hops several times, using one leg to move forward.", "segment_frame_end": 70, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "Strained_Hop_A03_001_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/g1/Strained_Hop_A03_001_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "sks person hops several times, using one leg to move forward.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00460.npz", "sample_index": 460}
{"caption": "sks person kick forward into empty space repeatedly.", "dataset": "permo_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 148830, "index": 1441, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00461.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/labels/Strained_Kick_A05_003_retarget__seg0000__part0000.json", "segment_action": "sks person kick forward into empty space repeatedly.", "segment_frame_end": 189, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "Strained_Kick_A05_003_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/g1/Strained_Kick_A05_003_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "sks person kick forward into empty space repeatedly.", "window_end": 189, "window_start": 129}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00461.npz", "sample_index": 461}
{"caption": "sks person is walking forward.", "dataset": "permo_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 148864, "index": 1475, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00462.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/labels/Strained_Walk_A04_001_retarget__seg0000__part0000.json", "segment_action": "sks person is walking forward.", "segment_frame_end": 147, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "Strained_Walk_A04_001_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/g1/Strained_Walk_A04_001_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "sks person is walking forward.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00462.npz", "sample_index": 462}
{"caption": "sks person waves with one arm repeatedly.", "dataset": "permo_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 148878, "index": 1489, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00463.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/labels/Strained_Wave_A05_003_retarget__seg0000__part0000.json", "segment_action": "sks person waves with one arm repeatedly.", "segment_frame_end": 103, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "Strained_Wave_A05_003_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/g1/Strained_Wave_A05_003_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "sks person waves with one arm repeatedly.", "window_end": 103, "window_start": 43}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00463.npz", "sample_index": 463}
{"caption": "sks person delivers repeated punches with one arm.", "dataset": "permo_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 148958, "index": 1569, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00464.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/labels/Teenage_Punch_A02_003_retarget__seg0000__part0000.json", "segment_action": "sks person delivers repeated punches with one arm.", "segment_frame_end": 127, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "Teenage_Punch_A02_003_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/g1/Teenage_Punch_A02_003_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "sks person delivers repeated punches with one arm.", "window_end": 94, "window_start": 34}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00464.npz", "sample_index": 464}
{"caption": "sks person walks in a straight line, moving forward.", "dataset": "permo_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 149148, "index": 1759, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00465.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/labels/Waiter_Walk_A04_002_retarget__seg0000__part0000.json", "segment_action": "sks person walks in a straight line, moving forward.", "segment_frame_end": 87, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "Waiter_Walk_A04_002_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/permo/g1/Waiter_Walk_A04_002_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "sks person walks in a straight line, moving forward.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00465.npz", "sample_index": 465}
{"caption": "The person is standing up straight, shaking their head slightly and moving their fingers.", "dataset": "snapmogen_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 149784, "index": 576, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00466.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/labels/ep1_00087#572#832_retarget__seg0000__part0000.json", "segment_action": "The person is standing up straight, shaking their head slightly and moving their fingers.", "segment_frame_end": 260, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "ep1_00087#572#832_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/g1/ep1_00087#572#832_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The person is standing up straight, shaking their head slightly and moving their fingers.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00466.npz", "sample_index": 466}
{"caption": "First, the person walks forward and stops. Then she waves her straight arms, turning her body to her right and her left. Then she turns to her left 180 degrees and walks forward. Finally, she turns 180 degrees to her right and stops.", "dataset": "snapmogen_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 149840, "index": 632, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00467.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/labels/ep1_00099#6504#6794_retarget__seg0000__part0000.json", "segment_action": "First, the person walks forward and stops. Then she waves her straight arms, turning her body to her right and her left. Then she turns to her left 180 degrees and walks forward. Finally, she turns 180 degrees to her right and stops.", "segment_frame_end": 290, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "ep1_00099#6504#6794_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/g1/ep1_00099#6504#6794_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "First, the person walks forward and stops. Then she waves her straight arms, turning her body to her right and her left. Then she turns to her left 180 degrees and walks forward. Finally, she turns 180 degrees to her right and stops.", "window_end": 290, "window_start": 230}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00467.npz", "sample_index": 467}
{"caption": "The person rehearses a waltz. First, turning slightly to the left, his right arm is bent at the elbow at chest level, his left arm is bent at the elbow at stomach level, his head is turned to the left, and he takes a step to the right, leaning slightly to the right, and straightens up. Then, he repeats this movement and turns 360 degrees. After that, he raises his right arm bent at the elbow to his neck, his left arm is straightened along the torso, and he turns 180 degrees to the left. At the end, he lowers his head down and looks at his palms, which are bent at the elbows at stomach level.", "dataset": "snapmogen_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 149866, "index": 658, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00468.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/labels/ep1_00107#1238#1493_retarget__seg0000__part0000.json", "segment_action": "The person rehearses a waltz. First, turning slightly to the left, his right arm is bent at the elbow at chest level, his left arm is bent at the elbow at stomach level, his head is turned to the left, and he takes a step to the right, leaning slightly to the right, and straightens up. Then, he repeats this movement and turns 360 degrees. After that, he raises his right arm bent at the elbow to his neck, his left arm is straightened along the torso, and he turns 180 degrees to the left. At the end, he lowers his head down and looks at his palms, which are bent at the elbows at stomach level.", "segment_frame_end": 255, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "ep1_00107#1238#1493_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/g1/ep1_00107#1238#1493_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The person rehearses a waltz. First, turning slightly to the left, his right arm is bent at the elbow at chest level, his left arm is bent at the elbow at stomach level, his head is turned to the left, and he takes a step to the right, leaning slightly to the right, and straightens up. Then, he repeats this movement and turns 360 degrees. After that, he raises his right arm bent at the elbow to his neck, his left arm is straightened along the torso, and he turns 180 degrees to the left. At the end, he lowers his head down and looks at his palms, which are bent at the elbows at stomach level.", "window_end": 158, "window_start": 98}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00468.npz", "sample_index": 468}
{"caption": "The person sits restlessly, turning their head to the right while moving their hands up and down near their hips, then abruptly turns their head to the left and stands up, lowering their hands. Then the person walks forward and turns their head to the right, then stops and turns their head slightly to the left, takes a step back and to the left diagonally and then continues to walk backward while turning slightly to the left.", "dataset": "snapmogen_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 149920, "index": 712, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00469.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/labels/ep1_00116#192#443_retarget__seg0000__part0000.json", "segment_action": "The person sits restlessly, turning their head to the right while moving their hands up and down near their hips, then abruptly turns their head to the left and stands up, lowering their hands. Then the person walks forward and turns their head to the right, then stops and turns their head slightly to the left, takes a step back and to the left diagonally and then continues to walk backward while turning slightly to the left.", "segment_frame_end": 251, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "ep1_00116#192#443_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/g1/ep1_00116#192#443_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The person sits restlessly, turning their head to the right while moving their hands up and down near their hips, then abruptly turns their head to the left and stands up, lowering their hands. Then the person walks forward and turns their head to the right, then stops and turns their head slightly to the left, takes a step back and to the left diagonally and then continues to walk backward while turning slightly to the left.", "window_end": 156, "window_start": 96}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00469.npz", "sample_index": 469}
{"caption": "The person stands up straight and waves his left hand slightly, moving his wrist. Then he steps forward with his right foot, spreads his arms to the sides and moves both hands, moving his wrists and turns his head to the right.", "dataset": "snapmogen_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 150016, "index": 808, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00470.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/labels/ep1_00124#4822#5108_retarget__seg0000__part0000.json", "segment_action": "The person stands up straight and waves his left hand slightly, moving his wrist. Then he steps forward with his right foot, spreads his arms to the sides and moves both hands, moving his wrists and turns his head to the right.", "segment_frame_end": 286, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "ep1_00124#4822#5108_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/g1/ep1_00124#4822#5108_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The person stands up straight and waves his left hand slightly, moving his wrist. Then he steps forward with his right foot, spreads his arms to the sides and moves both hands, moving his wrists and turns his head to the right.", "window_end": 286, "window_start": 226}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00470.npz", "sample_index": 470}
{"caption": "The person stands bent forward, left hand down and right on the waist. Then he sways his body left and right and straightens up, raising his right hand to the waist and then lowering it forward.", "dataset": "snapmogen_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 150108, "index": 900, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00471.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/labels/ep1_00136#5102#5291_retarget__seg0000__part0000.json", "segment_action": "The person stands bent forward, left hand down and right on the waist. Then he sways his body left and right and straightens up, raising his right hand to the waist and then lowering it forward.", "segment_frame_end": 189, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "ep1_00136#5102#5291_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/g1/ep1_00136#5102#5291_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The person stands bent forward, left hand down and right on the waist. Then he sways his body left and right and straightens up, raising his right hand to the waist and then lowering it forward.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00471.npz", "sample_index": 471}
{"caption": "The person stands with legs slightly apart, leaning slightly to the right, and turns his head that way. Then he shifts his weight to the left hip and takes a step forward with his right foot, extending his right hand to grab a bottle. He brings the bottle to his ear, shakes it, and lowers it in front of him, opening the cap with his left hand. Holding the cap in his left hand, he leans forward slightly, turns left, and pours the liquid with his right hand. Finally, he twists the bottle and places it back with his right hand.", "dataset": "snapmogen_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 150614, "index": 1406, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00472.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/labels/ep1_00205#7648#8023_retarget__seg0000__part0001.json", "segment_action": "The person stands with legs slightly apart, leaning slightly to the right, and turns his head that way. Then he shifts his weight to the left hip and takes a step forward with his right foot, extending his right hand to grab a bottle. He brings the bottle to his ear, shakes it, and lowers it in front of him, opening the cap with his left hand. Holding the cap in his left hand, he leans forward slightly, turns left, and pours the liquid with his right hand. Finally, he twists the bottle and places it back with his right hand.", "segment_frame_end": 75, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "ep1_00205#7648#8023_retarget__seg0000__part0001", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/g1/ep1_00205#7648#8023_retarget__seg0000__part0001.npz", "split": "test", "text_path": null, "video_summary": "The person stands with legs slightly apart, leaning slightly to the right, and turns his head that way. Then he shifts his weight to the left hip and takes a step forward with his right foot, extending his right hand to grab a bottle. He brings the bottle to his ear, shakes it, and lowers it in front of him, opening the cap with his left hand. Holding the cap in his left hand, he leans forward slightly, turns left, and pours the liquid with his right hand. Finally, he twists the bottle and places it back with his right hand.", "window_end": 68, "window_start": 8}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00472.npz", "sample_index": 472}
{"caption": "First, the person stands slightly leaning to his right, moving his lowered right hand away, and placing his left hand on his waist. Then the person places his left foot on the toe behind him. After this, the person goes to his left, lowering his hands, turning his head to his left and forward.", "dataset": "snapmogen_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 150978, "index": 1770, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00473.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/labels/ep1_00270#824#1047_retarget__seg0000__part0000.json", "segment_action": "First, the person stands slightly leaning to his right, moving his lowered right hand away, and placing his left hand on his waist. Then the person places his left foot on the toe behind him. After this, the person goes to his left, lowering his hands, turning his head to his left and forward.", "segment_frame_end": 223, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "ep1_00270#824#1047_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/g1/ep1_00270#824#1047_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "First, the person stands slightly leaning to his right, moving his lowered right hand away, and placing his left hand on his waist. Then the person places his left foot on the toe behind him. After this, the person goes to his left, lowering his hands, turning his head to his left and forward.", "window_end": 142, "window_start": 82}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00473.npz", "sample_index": 473}
{"caption": "The person stands in a natural pose, turns slightly to the right, and bends down, picking up an object, picks it up, and places it slightly to his left, then stands in a natural pose, bends down in front of him, and picks up a heavy object.", "dataset": "snapmogen_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 151247, "index": 2039, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00474.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/labels/ep2_00228#0#244_retarget__seg0000__part0000.json", "segment_action": "The person stands in a natural pose, turns slightly to the right, and bends down, picking up an object, picks it up, and places it slightly to his left, then stands in a natural pose, bends down in front of him, and picks up a heavy object.", "segment_frame_end": 244, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "ep2_00228#0#244_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/g1/ep2_00228#0#244_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The person stands in a natural pose, turns slightly to the right, and bends down, picking up an object, picks it up, and places it slightly to his left, then stands in a natural pose, bends down in front of him, and picks up a heavy object.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00474.npz", "sample_index": 474}
{"caption": "The person stands upright, turning their head to the right, then starts walking forward, turning their head forward, then turning to the right, walking forward relative to themselves and turning to the left and standing upright.", "dataset": "snapmogen_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 151352, "index": 2144, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00475.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/labels/ep2_00423#336#677_retarget__seg0000__part0000.json", "segment_action": "The person stands upright, turning their head to the right, then starts walking forward, turning their head forward, then turning to the right, walking forward relative to themselves and turning to the left and standing upright.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "ep2_00423#336#677_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/g1/ep2_00423#336#677_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The person stands upright, turning their head to the right, then starts walking forward, turning their head forward, then turning to the right, walking forward relative to themselves and turning to the left and standing upright.", "window_end": 300, "window_start": 240}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00475.npz", "sample_index": 475}
{"caption": "The person stands up straight, turns indifferently to the right side, and starts walking in a circle, moving clockwise.", "dataset": "snapmogen_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 151478, "index": 2270, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00476.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/labels/ep2_00687#0#179_retarget__seg0000__part0000.json", "segment_action": "The person stands up straight, turns indifferently to the right side, and starts walking in a circle, moving clockwise.", "segment_frame_end": 179, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "ep2_00687#0#179_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/g1/ep2_00687#0#179_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The person stands up straight, turns indifferently to the right side, and starts walking in a circle, moving clockwise.", "window_end": 179, "window_start": 119}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00476.npz", "sample_index": 476}
{"caption": "The person quickly runs forward diagonally to the right, slightly leaning his body forward, then slows down and stops. She then turns 180° to the left and walks forward diagonally to the right. She then turns 180° to the right and quickly runs forward again.", "dataset": "snapmogen_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 151512, "index": 2304, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00477.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/labels/ep2_00745#0#261_retarget__seg0000__part0000.json", "segment_action": "The person quickly runs forward diagonally to the right, slightly leaning his body forward, then slows down and stops. She then turns 180° to the left and walks forward diagonally to the right. She then turns 180° to the right and quickly runs forward again.", "segment_frame_end": 261, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "ep2_00745#0#261_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/g1/ep2_00745#0#261_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The person quickly runs forward diagonally to the right, slightly leaning his body forward, then slows down and stops. She then turns 180° to the left and walks forward diagonally to the right. She then turns 180° to the right and quickly runs forward again.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00477.npz", "sample_index": 477}
{"caption": "The person stands in a natural position. Then she turns her head and her body to the left and right.", "dataset": "snapmogen_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 151518, "index": 2310, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00478.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/labels/ep2_00766#0#134_retarget__seg0000__part0000.json", "segment_action": "The person stands in a natural position. Then she turns her head and her body to the left and right.", "segment_frame_end": 134, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "ep2_00766#0#134_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/g1/ep2_00766#0#134_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The person stands in a natural position. Then she turns her head and her body to the left and right.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00478.npz", "sample_index": 478}
{"caption": "The person confidently takes four steps forward, turning their head to the left at the end and stopping. The person then turns to the right and confidently moves forward, sharply turning their head to the left and then straight again. The person takes four steps and stops in place.", "dataset": "snapmogen_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 151590, "index": 2382, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00479.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/labels/ep2_01009#0#275_retarget__seg0000__part0000.json", "segment_action": "The person confidently takes four steps forward, turning their head to the left at the end and stopping. The person then turns to the right and confidently moves forward, sharply turning their head to the left and then straight again. The person takes four steps and stops in place.", "segment_frame_end": 275, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "ep2_01009#0#275_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/g1/ep2_01009#0#275_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The person confidently takes four steps forward, turning their head to the left at the end and stopping. The person then turns to the right and confidently moves forward, sharply turning their head to the left and then straight again. The person takes four steps and stops in place.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00479.npz", "sample_index": 479}
{"caption": "The person is standing with her hands in front of her face. She holds her right palm to her ear, her left palm to her mouth, leans slightly forward, and shifts from foot to foot. At the end the person straightens up.", "dataset": "snapmogen_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 151821, "index": 2613, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00480.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/labels/ep2_01594#179#350_retarget__seg0000__part0000.json", "segment_action": "The person is standing with her hands in front of her face. She holds her right palm to her ear, her left palm to her mouth, leans slightly forward, and shifts from foot to foot. At the end the person straightens up.", "segment_frame_end": 171, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "ep2_01594#179#350_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/g1/ep2_01594#179#350_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The person is standing with her hands in front of her face. She holds her right palm to her ear, her left palm to her mouth, leans slightly forward, and shifts from foot to foot. At the end the person straightens up.", "window_end": 171, "window_start": 111}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00480.npz", "sample_index": 480}
{"caption": "The person jumps elegantly waving their arms forward, backward, forward, backward, and forward again.", "dataset": "snapmogen_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 152552, "index": 3344, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00481.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/labels/gp_00181#545#809_retarget__seg0000__part0000.json", "segment_action": "The person jumps elegantly waving their arms forward, backward, forward, backward, and forward again.", "segment_frame_end": 264, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "gp_00181#545#809_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/g1/gp_00181#545#809_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The person jumps elegantly waving their arms forward, backward, forward, backward, and forward again.", "window_end": 264, "window_start": 204}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00481.npz", "sample_index": 481}
{"caption": "The person walks forward with their legs wide apart and holding something with both hands in front of the stomach. They leaning their head to their left and then turn their head straight. They constantly keep their hands in front of them at stomach level.", "dataset": "snapmogen_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 152753, "index": 3545, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00482.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/labels/gp_00257#407#613_retarget__seg0000__part0000.json", "segment_action": "The person walks forward with their legs wide apart and holding something with both hands in front of the stomach. They leaning their head to their left and then turn their head straight. They constantly keep their hands in front of them at stomach level.", "segment_frame_end": 206, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "gp_00257#407#613_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/g1/gp_00257#407#613_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The person walks forward with their legs wide apart and holding something with both hands in front of the stomach. They leaning their head to their left and then turn their head straight. They constantly keep their hands in front of them at stomach level.", "window_end": 206, "window_start": 146}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00482.npz", "sample_index": 482}
{"caption": "A person stands in a neutral resting position. Then suddenly they stick their left arm out to the side and start waving it, bending it at the elbow. The person lowers their left arm to repeat it all over again.", "dataset": "snapmogen_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 153283, "index": 4075, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00483.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/labels/gp_00457#0#275_retarget__seg0000__part0000.json", "segment_action": "A person stands in a neutral resting position. Then suddenly they stick their left arm out to the side and start waving it, bending it at the elbow. The person lowers their left arm to repeat it all over again.", "segment_frame_end": 275, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "gp_00457#0#275_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/g1/gp_00457#0#275_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A person stands in a neutral resting position. Then suddenly they stick their left arm out to the side and start waving it, bending it at the elbow. The person lowers their left arm to repeat it all over again.", "window_end": 168, "window_start": 108}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00483.npz", "sample_index": 483}
{"caption": "A person with a calm expression stands slightly turned to the left with arms bent upward at the level of the neck and shakes their hands turning to the right, lowering their hands down and again raising their arms up, two times waving their hands up and down and again lowered slightly turning to the left, and finally repeats the first movement shaking their hands at the level of the neck.", "dataset": "snapmogen_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 153487, "index": 4279, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00484.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/labels/gp_00528#554#714_retarget__seg0000__part0000.json", "segment_action": "A person with a calm expression stands slightly turned to the left with arms bent upward at the level of the neck and shakes their hands turning to the right, lowering their hands down and again raising their arms up, two times waving their hands up and down and again lowered slightly turning to the left, and finally repeats the first movement shaking their hands at the level of the neck.", "segment_frame_end": 160, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "gp_00528#554#714_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/g1/gp_00528#554#714_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A person with a calm expression stands slightly turned to the left with arms bent upward at the level of the neck and shakes their hands turning to the right, lowering their hands down and again raising their arms up, two times waving their hands up and down and again lowered slightly turning to the left, and finally repeats the first movement shaking their hands at the level of the neck.", "window_end": 110, "window_start": 50}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00484.npz", "sample_index": 484}
{"caption": "The person standing with their right side extended their right arm forward, then lowered and turned 180 degrees over their right shoulder. The person stands neutral and then raises their arms to their chest and moves them confidently to the sides while turning slightly to the right.", "dataset": "snapmogen_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 153552, "index": 4344, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00485.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/labels/gp_00550#1204#1388_retarget__seg0000__part0000.json", "segment_action": "The person standing with their right side extended their right arm forward, then lowered and turned 180 degrees over their right shoulder. The person stands neutral and then raises their arms to their chest and moves them confidently to the sides while turning slightly to the right.", "segment_frame_end": 184, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "gp_00550#1204#1388_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/g1/gp_00550#1204#1388_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The person standing with their right side extended their right arm forward, then lowered and turned 180 degrees over their right shoulder. The person stands neutral and then raises their arms to their chest and moves them confidently to the sides while turning slightly to the right.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00485.npz", "sample_index": 485}
{"caption": "The person is standing in a boxing stance with their feet wide apart and their torso turned to the left, holding their right arm outstretched in front of them. The person turns their torso to the right and brings their right arm toward them, shaking their head. Then they turn to the right and strike forward with their left hand, turn to the left and extend their left arm toward them. The person shakes their body, lowers their arms and takes a step forward", "dataset": "snapmogen_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 153995, "index": 4787, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00486.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/labels/m_ep1_00007#173#392_retarget__seg0000__part0000.json", "segment_action": "The person is standing in a boxing stance with their feet wide apart and their torso turned to the left, holding their right arm outstretched in front of them. The person turns their torso to the right and brings their right arm toward them, shaking their head. Then they turn to the right and strike forward with their left hand, turn to the left and extend their left arm toward them. The person shakes their body, lowers their arms and takes a step forward", "segment_frame_end": 219, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "m_ep1_00007#173#392_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/g1/m_ep1_00007#173#392_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The person is standing in a boxing stance with their feet wide apart and their torso turned to the left, holding their right arm outstretched in front of them. The person turns their torso to the right and brings their right arm toward them, shaking their head. Then they turn to the right and strike forward with their left hand, turn to the left and extend their left arm toward them. The person shakes their body, lowers their arms and takes a step forward", "window_end": 140, "window_start": 80}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00486.npz", "sample_index": 486}
{"caption": "The person completes the stretch. Standing with legs wide apart, the person stretches her left leg sideways, transferring her weight to her bent right leg. The person then straightens up onto her straight legs, stands on her left leg, and then swings it freely with her arms.", "dataset": "snapmogen_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 154019, "index": 4811, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00487.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/labels/m_ep1_00009#4629#4896_retarget__seg0000__part0000.json", "segment_action": "The person completes the stretch. Standing with legs wide apart, the person stretches her left leg sideways, transferring her weight to her bent right leg. The person then straightens up onto her straight legs, stands on her left leg, and then swings it freely with her arms.", "segment_frame_end": 267, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "m_ep1_00009#4629#4896_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/g1/m_ep1_00009#4629#4896_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The person completes the stretch. Standing with legs wide apart, the person stretches her left leg sideways, transferring her weight to her bent right leg. The person then straightens up onto her straight legs, stands on her left leg, and then swings it freely with her arms.", "window_end": 164, "window_start": 104}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00487.npz", "sample_index": 487}
{"caption": "The person stands in a natural pose and then grasps his neck on the left side with his left hand, then tilts his torso slightly to the right, takes a small step forward with his left foot, and waves his left arm up and down in front of him.", "dataset": "snapmogen_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 154608, "index": 5400, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00488.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/labels/m_ep1_00096#2547#2715_retarget__seg0000__part0000.json", "segment_action": "The person stands in a natural pose and then grasps his neck on the left side with his left hand, then tilts his torso slightly to the right, takes a small step forward with his left foot, and waves his left arm up and down in front of him.", "segment_frame_end": 168, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "m_ep1_00096#2547#2715_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/g1/m_ep1_00096#2547#2715_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The person stands in a natural pose and then grasps his neck on the left side with his left hand, then tilts his torso slightly to the right, takes a small step forward with his left foot, and waves his left arm up and down in front of him.", "window_end": 168, "window_start": 108}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00488.npz", "sample_index": 488}
{"caption": "The person stands in a neutral resting position, then lowers her head and takes a step backward. After that, she turns right and walks forward. Finally, the person stops, turns left and stands in a neutral position.", "dataset": "snapmogen_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 154665, "index": 5457, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00489.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/labels/m_ep1_00110#1266#1517_retarget__seg0000__part0000.json", "segment_action": "The person stands in a neutral resting position, then lowers her head and takes a step backward. After that, she turns right and walks forward. Finally, the person stops, turns left and stands in a neutral position.", "segment_frame_end": 251, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "m_ep1_00110#1266#1517_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/g1/m_ep1_00110#1266#1517_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The person stands in a neutral resting position, then lowers her head and takes a step backward. After that, she turns right and walks forward. Finally, the person stops, turns left and stands in a neutral position.", "window_end": 251, "window_start": 191}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00489.npz", "sample_index": 489}
{"caption": "The person stands straight and with her hands together near her stomach, after which she jumps up and spreads her legs to the sides and raises her arms and brings them together, slightly bent at the elbows. Then she jumps again and returns to her original position. Then she jumps again, spreads her legs to the sides, raises her arms, bent at the elbows, and then brings her legs together, initially the right one, then the left one, while the right hand, bent at the elbow, touches the stomach and the left hand, touches the left cheek at this time the head is slightly tilted to the left.", "dataset": "snapmogen_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 155275, "index": 6067, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00490.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/labels/m_ep1_00199#3079#3262_retarget__seg0000__part0000.json", "segment_action": "The person stands straight and with her hands together near her stomach, after which she jumps up and spreads her legs to the sides and raises her arms and brings them together, slightly bent at the elbows. Then she jumps again and returns to her original position. Then she jumps again, spreads her legs to the sides, raises her arms, bent at the elbows, and then brings her legs together, initially the right one, then the left one, while the right hand, bent at the elbow, touches the stomach and the left hand, touches the left cheek at this time the head is slightly tilted to the left.", "segment_frame_end": 183, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "m_ep1_00199#3079#3262_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/g1/m_ep1_00199#3079#3262_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The person stands straight and with her hands together near her stomach, after which she jumps up and spreads her legs to the sides and raises her arms and brings them together, slightly bent at the elbows. Then she jumps again and returns to her original position. Then she jumps again, spreads her legs to the sides, raises her arms, bent at the elbows, and then brings her legs together, initially the right one, then the left one, while the right hand, bent at the elbow, touches the stomach and the left hand, touches the left cheek at this time the head is slightly tilted to the left.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00490.npz", "sample_index": 490}
{"caption": "The person stands restlessly, moving the left foot closer to the right and leaning slightly to the left while the arms are lowered. Then the person moves the right leg to the right while shifting the body weight to the right leg and folds the hands together in front of the hips. After that, the person steps back, lowers their arms, and takes two steps forward while slightly raising and lowering their left arm and turning their head to the right. Then the person raises their hands to the level of their stomach and gestures with them as if they were talking to someone.", "dataset": "snapmogen_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 155550, "index": 6342, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00491.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/labels/m_ep1_00233#842#1127_retarget__seg0000__part0000.json", "segment_action": "The person stands restlessly, moving the left foot closer to the right and leaning slightly to the left while the arms are lowered. Then the person moves the right leg to the right while shifting the body weight to the right leg and folds the hands together in front of the hips. After that, the person steps back, lowers their arms, and takes two steps forward while slightly raising and lowering their left arm and turning their head to the right. Then the person raises their hands to the level of their stomach and gestures with them as if they were talking to someone.", "segment_frame_end": 285, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "m_ep1_00233#842#1127_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/g1/m_ep1_00233#842#1127_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The person stands restlessly, moving the left foot closer to the right and leaning slightly to the left while the arms are lowered. Then the person moves the right leg to the right while shifting the body weight to the right leg and folds the hands together in front of the hips. After that, the person steps back, lowers their arms, and takes two steps forward while slightly raising and lowering their left arm and turning their head to the right. Then the person raises their hands to the level of their stomach and gestures with them as if they were talking to someone.", "window_end": 172, "window_start": 112}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00491.npz", "sample_index": 491}
{"caption": "The person stands with his right leg slightly raised to the side and arms outstretched, then straightens up. Then, he jumps sideways to his left three times with small jumps. Turns his head to the left for a moment, turns his right 90 degrees, and goes forward.", "dataset": "snapmogen_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 155818, "index": 6610, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00492.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/labels/m_ep1_00280#5084#5331_retarget__seg0000__part0000.json", "segment_action": "The person stands with his right leg slightly raised to the side and arms outstretched, then straightens up. Then, he jumps sideways to his left three times with small jumps. Turns his head to the left for a moment, turns his right 90 degrees, and goes forward.", "segment_frame_end": 247, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "m_ep1_00280#5084#5331_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/g1/m_ep1_00280#5084#5331_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The person stands with his right leg slightly raised to the side and arms outstretched, then straightens up. Then, he jumps sideways to his left three times with small jumps. Turns his head to the left for a moment, turns his right 90 degrees, and goes forward.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00492.npz", "sample_index": 492}
{"caption": "The person stands in a neutral position, then walks forward energetically, starting with their right foot, swinging their arms back and forth. After this, the person stops in a neutral position, turning their body slightly to their right.", "dataset": "snapmogen_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 155885, "index": 6677, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00493.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/labels/m_ep1_00291#2968#3285_retarget__seg0000__part0000.json", "segment_action": "The person stands in a neutral position, then walks forward energetically, starting with their right foot, swinging their arms back and forth. After this, the person stops in a neutral position, turning their body slightly to their right.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "m_ep1_00291#2968#3285_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/g1/m_ep1_00291#2968#3285_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The person stands in a neutral position, then walks forward energetically, starting with their right foot, swinging their arms back and forth. After this, the person stops in a neutral position, turning their body slightly to their right.", "window_end": 180, "window_start": 120}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00493.npz", "sample_index": 493}
{"caption": "The person lunges forward with the right foot and squats on one leg, rests both hands on the right thigh, stands in a natural pose, and lunges with the right foot two more times.", "dataset": "snapmogen_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 156139, "index": 6931, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00494.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/labels/m_ep2_00389#0#335_retarget__seg0000__part0000.json", "segment_action": "The person lunges forward with the right foot and squats on one leg, rests both hands on the right thigh, stands in a natural pose, and lunges with the right foot two more times.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "m_ep2_00389#0#335_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/g1/m_ep2_00389#0#335_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The person lunges forward with the right foot and squats on one leg, rests both hands on the right thigh, stands in a natural pose, and lunges with the right foot two more times.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00494.npz", "sample_index": 494}
{"caption": "A person walks calmly in a semi-oval trajectory, turning to the right. At the end, they stop and turn to the right.", "dataset": "snapmogen_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 156482, "index": 7274, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00495.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/labels/m_ep2_01256#138#308_retarget__seg0000__part0000.json", "segment_action": "A person walks calmly in a semi-oval trajectory, turning to the right. At the end, they stop and turn to the right.", "segment_frame_end": 170, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "m_ep2_01256#138#308_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/g1/m_ep2_01256#138#308_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A person walks calmly in a semi-oval trajectory, turning to the right. At the end, they stop and turn to the right.", "window_end": 170, "window_start": 110}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00495.npz", "sample_index": 495}
{"caption": "The person walks in a circle. He is in a neutral resting position with his arms down. He then begins to walk, turning back over his right shoulder in a circular path and waving his arms slightly next to his body.", "dataset": "snapmogen_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 156769, "index": 7561, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00496.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/labels/m_ep2_01944#0#236_retarget__seg0000__part0000.json", "segment_action": "The person walks in a circle. He is in a neutral resting position with his arms down. He then begins to walk, turning back over his right shoulder in a circular path and waving his arms slightly next to his body.", "segment_frame_end": 236, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "m_ep2_01944#0#236_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/g1/m_ep2_01944#0#236_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The person walks in a circle. He is in a neutral resting position with his arms down. He then begins to walk, turning back over his right shoulder in a circular path and waving his arms slightly next to his body.", "window_end": 148, "window_start": 88}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00496.npz", "sample_index": 496}
{"caption": "The person jumps forward and then backward, then left and right with their arms raised to their chest.", "dataset": "snapmogen_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 156837, "index": 7629, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00497.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/labels/m_ep2_02062#0#135_retarget__seg0000__part0000.json", "segment_action": "The person jumps forward and then backward, then left and right with their arms raised to their chest.", "segment_frame_end": 135, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "m_ep2_02062#0#135_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/g1/m_ep2_02062#0#135_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The person jumps forward and then backward, then left and right with their arms raised to their chest.", "window_end": 98, "window_start": 38}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00497.npz", "sample_index": 497}
{"caption": "The person sits and holds a knife in their right hand and a fork in their left hand and cuts food by raising their hand with the fork to their mouth and raising their head.", "dataset": "snapmogen_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 157206, "index": 7998, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00498.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/labels/m_gp_00149#286#506_retarget__seg0000__part0000.json", "segment_action": "The person sits and holds a knife in their right hand and a fork in their left hand and cuts food by raising their hand with the fork to their mouth and raising their head.", "segment_frame_end": 220, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "m_gp_00149#286#506_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/g1/m_gp_00149#286#506_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The person sits and holds a knife in their right hand and a fork in their left hand and cuts food by raising their hand with the fork to their mouth and raising their head.", "window_end": 140, "window_start": 80}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00498.npz", "sample_index": 498}
{"caption": "The person sits, moves his left bent hand to his mouth, and then moves it back down. His right hand is on his right leg.", "dataset": "snapmogen_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 157730, "index": 8522, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00499.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/labels/m_gp_00344#425#709_retarget__seg0000__part0000.json", "segment_action": "The person sits, moves his left bent hand to his mouth, and then moves it back down. His right hand is on his right leg.", "segment_frame_end": 284, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "m_gp_00344#425#709_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/g1/m_gp_00344#425#709_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The person sits, moves his left bent hand to his mouth, and then moves it back down. His right hand is on his right leg.", "window_end": 172, "window_start": 112}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00499.npz", "sample_index": 499}
{"caption": "A person stands neutrally and then walks forward and stands on their toes while stretching their arms up, then jumps up and lowers their arms a little, and then steps back while swinging their arms a little.", "dataset": "snapmogen_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 157897, "index": 8689, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00500.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/labels/m_gp_00418#2489#2872_retarget__seg0000__part0000.json", "segment_action": "A person stands neutrally and then walks forward and stands on their toes while stretching their arms up, then jumps up and lowers their arms a little, and then steps back while swinging their arms a little.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "m_gp_00418#2489#2872_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/g1/m_gp_00418#2489#2872_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A person stands neutrally and then walks forward and stands on their toes while stretching their arms up, then jumps up and lowers their arms a little, and then steps back while swinging their arms a little.", "window_end": 300, "window_start": 240}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00500.npz", "sample_index": 500}
{"caption": "The person is dynamically stepping on the box. The person climbs off the top of the box with their left leg, then turns to their left and climbs up on the box with their left leg.", "dataset": "snapmogen_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 158033, "index": 8825, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00501.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/labels/m_gp_00452#617#917_retarget__seg0000__part0000.json", "segment_action": "The person is dynamically stepping on the box. The person climbs off the top of the box with their left leg, then turns to their left and climbs up on the box with their left leg.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "m_gp_00452#617#917_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/g1/m_gp_00452#617#917_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The person is dynamically stepping on the box. The person climbs off the top of the box with their left leg, then turns to their left and climbs up on the box with their left leg.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00501.npz", "sample_index": 501}
{"caption": "The person stands swaying to the right and left, holding their bent arms together and scratching them. The head is turned to the right, lowered down and turned to the left, as if looking around, then lowered and stood upleft.", "dataset": "snapmogen_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 158079, "index": 8871, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00502.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/labels/m_gp_00470#124#424_retarget__seg0000__part0000.json", "segment_action": "The person stands swaying to the right and left, holding their bent arms together and scratching them. The head is turned to the right, lowered down and turned to the left, as if looking around, then lowered and stood upleft.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "m_gp_00470#124#424_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/g1/m_gp_00470#124#424_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The person stands swaying to the right and left, holding their bent arms together and scratching them. The head is turned to the right, lowered down and turned to the left, as if looking around, then lowered and stood upleft.", "window_end": 180, "window_start": 120}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00502.npz", "sample_index": 502}
{"caption": "A person stands neutrally, then bends their right arm and raises it near their hip, and with their left hand they wind something over their right arm, purs a liquid with the right hand over the left hand. All the while, the legs are straight.", "dataset": "snapmogen_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 158111, "index": 8903, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00503.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/labels/m_gp_00479#0#190_retarget__seg0000__part0000.json", "segment_action": "A person stands neutrally, then bends their right arm and raises it near their hip, and with their left hand they wind something over their right arm, purs a liquid with the right hand over the left hand. All the while, the legs are straight.", "segment_frame_end": 190, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "m_gp_00479#0#190_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/g1/m_gp_00479#0#190_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A person stands neutrally, then bends their right arm and raises it near their hip, and with their left hand they wind something over their right arm, purs a liquid with the right hand over the left hand. All the while, the legs are straight.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00503.npz", "sample_index": 503}
{"caption": "A person is standing and bows twice. During the first bow, he has his elbows bent and his hands together in front of his chest. He covers his right fist with his open left palm. During the second bow, he lowers his arms and places his hands on his thighs.", "dataset": "snapmogen_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 158488, "index": 9280, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00504.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/labels/m_gp_00613#1208#1378_retarget__seg0000__part0000.json", "segment_action": "A person is standing and bows twice. During the first bow, he has his elbows bent and his hands together in front of his chest. He covers his right fist with his open left palm. During the second bow, he lowers his arms and places his hands on his thighs.", "segment_frame_end": 170, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "m_gp_00613#1208#1378_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/g1/m_gp_00613#1208#1378_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A person is standing and bows twice. During the first bow, he has his elbows bent and his hands together in front of his chest. He covers his right fist with his open left palm. During the second bow, he lowers his arms and places his hands on his thighs.", "window_end": 115, "window_start": 55}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00504.npz", "sample_index": 504}
{"caption": "The person walks forward, relaxing with his head tilted slightly down. Then he stands in a neutral resting position facing forth.", "dataset": "snapmogen_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 158656, "index": 9448, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00505.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/labels/m_gp_00668#1076#1257_retarget__seg0000__part0000.json", "segment_action": "The person walks forward, relaxing with his head tilted slightly down. Then he stands in a neutral resting position facing forth.", "segment_frame_end": 181, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "m_gp_00668#1076#1257_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/g1/m_gp_00668#1076#1257_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The person walks forward, relaxing with his head tilted slightly down. Then he stands in a neutral resting position facing forth.", "window_end": 120, "window_start": 60}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00505.npz", "sample_index": 505}
{"caption": "The person is standing on bent legs in fear and leaning forward while holding their right hand on their right knee and their left hand raised to chest level. The person then leans back, raising their left hand to their face, then leans forward again, raising their hands to their chest. The person then turns 180 degrees to the left and runs forward in a hurry, then stops and leans forward again. The person sways slightly while keeping their hands at the level of the stomach. Finally, the person turns to the left while straightening up.", "dataset": "snapmogen_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 158784, "index": 9576, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00506.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/labels/m_st2_00001#265#562_retarget__seg0000__part0000.json", "segment_action": "The person is standing on bent legs in fear and leaning forward while holding their right hand on their right knee and their left hand raised to chest level. The person then leans back, raising their left hand to their face, then leans forward again, raising their hands to their chest. The person then turns 180 degrees to the left and runs forward in a hurry, then stops and leans forward again. The person sways slightly while keeping their hands at the level of the stomach. Finally, the person turns to the left while straightening up.", "segment_frame_end": 297, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "m_st2_00001#265#562_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/g1/m_st2_00001#265#562_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The person is standing on bent legs in fear and leaning forward while holding their right hand on their right knee and their left hand raised to chest level. The person then leans back, raising their left hand to their face, then leans forward again, raising their hands to their chest. The person then turns 180 degrees to the left and runs forward in a hurry, then stops and leans forward again. The person sways slightly while keeping their hands at the level of the stomach. Finally, the person turns to the left while straightening up.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00506.npz", "sample_index": 506}
{"caption": "The person walks forward like a zombie with their legs wide apart, sways their body back and forth, and raises their right arm to shoulder height. The person then bends down and pulls the right arm down and raises, lowering and raising the head.", "dataset": "snapmogen_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 158821, "index": 9613, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00507.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/labels/m_st2_00030#0#158_retarget__seg0000__part0000.json", "segment_action": "The person walks forward like a zombie with their legs wide apart, sways their body back and forth, and raises their right arm to shoulder height. The person then bends down and pulls the right arm down and raises, lowering and raising the head.", "segment_frame_end": 158, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "m_st2_00030#0#158_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/g1/m_st2_00030#0#158_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The person walks forward like a zombie with their legs wide apart, sways their body back and forth, and raises their right arm to shoulder height. The person then bends down and pulls the right arm down and raises, lowering and raising the head.", "window_end": 109, "window_start": 49}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00507.npz", "sample_index": 507}
{"caption": "A person walks in a semicircle, smoothly turning to the right, and holds a leash in his left hand.", "dataset": "snapmogen_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 158982, "index": 9774, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00508.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 2, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/labels/m_st2_00216#257#509_retarget__seg0000__part0000.json", "segment_action": "A person walks in a semicircle, smoothly turning to the right, and holds a leash in his left hand.", "segment_frame_end": 252, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "m_st2_00216#257#509_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/g1/m_st2_00216#257#509_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "A person walks in a semicircle, smoothly turning to the right, and holds a leash in his left hand.", "window_end": 252, "window_start": 192}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00508.npz", "sample_index": 508}
{"caption": "The person stands leaning on their left leg, slightly tilted to their left side. They slowly raise their left hand upwards to chin level, opening and closing the palm several times before lowering it down. Then, they extend their right hand forward to chin level, opening the palm, and lower it down. Next, they extend the left hand again to chin level, opening the palm twice. Finally, they stand upleft, extend the right hand forward to chin level, open the palm, and lower the hand down.", "dataset": "snapmogen_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 159062, "index": 9854, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00509.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/labels/m_st3_00027#170#472_retarget__seg0000__part0000.json", "segment_action": "The person stands leaning on their left leg, slightly tilted to their left side. They slowly raise their left hand upwards to chin level, opening and closing the palm several times before lowering it down. Then, they extend their right hand forward to chin level, opening the palm, and lower it down. Next, they extend the left hand again to chin level, opening the palm twice. Finally, they stand upleft, extend the right hand forward to chin level, open the palm, and lower the hand down.", "segment_frame_end": 300, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "m_st3_00027#170#472_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/g1/m_st3_00027#170#472_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The person stands leaning on their left leg, slightly tilted to their left side. They slowly raise their left hand upwards to chin level, opening and closing the palm several times before lowering it down. Then, they extend their right hand forward to chin level, opening the palm, and lower it down. Next, they extend the left hand again to chin level, opening the palm twice. Finally, they stand upleft, extend the right hand forward to chin level, open the palm, and lower the hand down.", "window_end": 180, "window_start": 120}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00509.npz", "sample_index": 509}
{"caption": "The person turns slightly to the right, tilting their body forward and raising their left hand to their face. The person turns to the left and moves the arms back to a neutral position.", "dataset": "snapmogen_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 160957, "index": 11749, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00510.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 0, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/labels/st3_00587#508#750_retarget__seg0000__part0000.json", "segment_action": "The person turns slightly to the right, tilting their body forward and raising their left hand to their face. The person turns to the left and moves the arms back to a neutral position.", "segment_frame_end": 242, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "st3_00587#508#750_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/g1/st3_00587#508#750_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The person turns slightly to the right, tilting their body forward and raising their left hand to their face. The person turns to the left and moves the arms back to a neutral position.", "window_end": 60, "window_start": 0}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00510.npz", "sample_index": 510}
{"caption": "The person stands in a neutral resting pose leaning to the right, crossing their arms in the stomach area. Bending and straightening their right arm at the elbow several times, the person moves their left arm up and down a little. They lower their arms, straighten up, bend and raise their head, and turn them to the left.", "dataset": "snapmogen_original_filtered_test", "fps": 30.0, "frames": 60, "global_index": 161172, "index": 11964, "keypoints_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00511.npz", "meta": {"eval_num_windows": 3, "eval_window_index": 1, "fps": 30.0, "label_path": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/labels/st3_00782#567#704_retarget__seg0000__part0000.json", "segment_action": "The person stands in a neutral resting pose leaning to the right, crossing their arms in the stomach area. Bending and straightening their right arm at the elbow several times, the person moves their left arm up and down a little. They lower their arms, straighten up, bend and raise their head, and turn them to the left.", "segment_frame_end": 137, "segment_frame_start": 0, "segment_index": 0, "segment_style": "", "sequence_name": "st3_00782#567#704_retarget__seg0000__part0000", "source_file": "${OMG_STORAGE_ROOT}/filtered_datasets/original/snapmogen/g1/st3_00782#567#704_retarget__seg0000__part0000.npz", "split": "test", "text_path": null, "video_summary": "The person stands in a neutral resting pose leaning to the right, crossing their arms in the stomach area. Bending and straightening their right arm at the elbow several times, the person moves their left arm up and down a little. They lower their arms, straighten up, bend and raise their head, and turn them to the left.", "window_end": 98, "window_start": 38}, "motion_path": "${OMG_REPO_ROOT}/outputs/benchmark_samples/mixed_modalities_all_v1/humanref_inputs/keypoints/sample_00511.npz", "sample_index": 511}