upload subset_20k/test_100 's json, videos and frames

data/subset_20k/test_100/data_worsen_gen.json

@@ -0,0 +1,336 @@
+[
+    {
+        "id": "b204331",
+        "images": [
+            "b204331_00.jpg",
+            "b204331_01.jpg",
+            "b204331_02.jpg",
+            "b204331_03.jpg",
+            "b204331_04.jpg",
+            "b204331_05.jpg",
+            "b204331_06.jpg",
+            "b204331_07.jpg",
+            "b204331_08.jpg",
+            "b204331_09.jpg",
+            "b204331_10.jpg",
+            "b204331_11.jpg",
+            "b204331_12.jpg",
+            "b204331_13.jpg",
+            "b204331_14.jpg",
+            "b204331_15.jpg"
+        ],
+        "conversations": [
+            {
+                "from": "human",
+                "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is the cell donates an empty electron to an oxygen molecule  ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> \n"
+            },
+            {
+                "from": "gpt",
+                "value": "visual quality: 3\n object consistency: 3\n dynamic degree: 1\n motion smoothness: 3\n text-to-video alignment: 1\n factual consistency: 3\n overall score: 1\n"
+            }
+        ]
+    },
+    {
+        "id": "b604211",
+        "images": [
+            "b604211_00.jpg",
+            "b604211_01.jpg",
+            "b604211_02.jpg",
+            "b604211_03.jpg",
+            "b604211_04.jpg",
+            "b604211_05.jpg",
+            "b604211_06.jpg",
+            "b604211_07.jpg",
+            "b604211_08.jpg",
+            "b604211_09.jpg",
+            "b604211_10.jpg",
+            "b604211_11.jpg",
+            "b604211_12.jpg",
+            "b604211_13.jpg",
+            "b604211_14.jpg",
+            "b604211_15.jpg"
+        ],
+        "conversations": [
+            {
+                "from": "human",
+                "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is Create a vignette for a course whose theme is the importance of women in the job market where the title is Plural Voices: Women  ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> \n"
+            },
+            {
+                "from": "gpt",
+                "value": "visual quality: 1\n object consistency: 3\n dynamic degree: 1\n motion smoothness: 1\n text-to-video alignment: 1\n factual consistency: 3\n overall score: 1\n"
+            }
+        ]
+    },
+    {
+        "id": "b205316",
+        "images": [
+            "b205316_00.jpg",
+            "b205316_01.jpg",
+            "b205316_02.jpg",
+            "b205316_03.jpg",
+            "b205316_04.jpg",
+            "b205316_05.jpg",
+            "b205316_06.jpg",
+            "b205316_07.jpg",
+            "b205316_08.jpg",
+            "b205316_09.jpg",
+            "b205316_10.jpg",
+            "b205316_11.jpg",
+            "b205316_12.jpg",
+            "b205316_13.jpg",
+            "b205316_14.jpg",
+            "b205316_15.jpg"
+        ],
+        "conversations": [
+            {
+                "from": "human",
+                "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is absolute reality stunning photo of a woman in cashmere, loose fitting, casual, black sweater with long sleeves on crosswalk in New York City  ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> \n"
+            },
+            {
+                "from": "gpt",
+                "value": "visual quality: 1\n object consistency: 3\n dynamic degree: 1\n motion smoothness: 3\n text-to-video alignment: 3\n factual consistency: 2\n overall score: 1\n"
+            }
+        ]
+    },
+    {
+        "id": "b802489",
+        "images": [
+            "b802489_00.jpg",
+            "b802489_01.jpg",
+            "b802489_02.jpg",
+            "b802489_03.jpg",
+            "b802489_04.jpg",
+            "b802489_05.jpg",
+            "b802489_06.jpg",
+            "b802489_07.jpg",
+            "b802489_08.jpg",
+            "b802489_09.jpg",
+            "b802489_10.jpg",
+            "b802489_11.jpg",
+            "b802489_12.jpg",
+            "b802489_13.jpg",
+            "b802489_14.jpg",
+            "b802489_15.jpg",
+            "b802489_16.jpg",
+            "b802489_17.jpg",
+            "b802489_18.jpg",
+            "b802489_19.jpg",
+            "b802489_20.jpg",
+            "b802489_21.jpg",
+            "b802489_22.jpg",
+            "b802489_23.jpg"
+        ],
+        "conversations": [
+            {
+                "from": "human",
+                "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is warrior raising sword in the rain.  ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> \n"
+            },
+            {
+                "from": "gpt",
+                "value": "visual quality: 1\n object consistency: 3\n dynamic degree: 1\n motion smoothness: 2\n text-to-video alignment: 1\n factual consistency: 3\n overall score: 1\n"
+            }
+        ]
+    },
+    {
+        "id": "b802941",
+        "images": [
+            "b802941_00.jpg",
+            "b802941_01.jpg",
+            "b802941_02.jpg",
+            "b802941_03.jpg",
+            "b802941_04.jpg",
+            "b802941_05.jpg",
+            "b802941_06.jpg",
+            "b802941_07.jpg",
+            "b802941_08.jpg",
+            "b802941_09.jpg",
+            "b802941_10.jpg",
+            "b802941_11.jpg",
+            "b802941_12.jpg",
+            "b802941_13.jpg",
+            "b802941_14.jpg",
+            "b802941_15.jpg",
+            "b802941_16.jpg",
+            "b802941_17.jpg",
+            "b802941_18.jpg",
+            "b802941_19.jpg",
+            "b802941_20.jpg",
+            "b802941_21.jpg",
+            "b802941_22.jpg",
+            "b802941_23.jpg"
+        ],
+        "conversations": [
+            {
+                "from": "human",
+                "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is claymation weird human midsommar mask covreing face, high detail, lots of detail, sunset, colourful, weird, creepy, surreal, macro detail  ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> \n"
+            },
+            {
+                "from": "gpt",
+                "value": "visual quality: 1\n object consistency: 1\n dynamic degree: 2\n motion smoothness: 3\n text-to-video alignment: 2\n factual consistency: 3\n overall score: 1\n"
+            }
+        ]
+    },
+    {
+        "id": "b602727",
+        "images": [
+            "b602727_00.jpg",
+            "b602727_01.jpg",
+            "b602727_02.jpg",
+            "b602727_03.jpg",
+            "b602727_04.jpg",
+            "b602727_05.jpg",
+            "b602727_06.jpg",
+            "b602727_07.jpg",
+            "b602727_08.jpg",
+            "b602727_09.jpg",
+            "b602727_10.jpg",
+            "b602727_11.jpg",
+            "b602727_12.jpg",
+            "b602727_13.jpg",
+            "b602727_14.jpg",
+            "b602727_15.jpg"
+        ],
+        "conversations": [
+            {
+                "from": "human",
+                "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is drummer of tool plays on piano  ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> \n"
+            },
+            {
+                "from": "gpt",
+                "value": "visual quality: 2\n object consistency: 3\n dynamic degree: 1\n motion smoothness: 1\n text-to-video alignment: 2\n factual consistency: 3\n overall score: 1\n"
+            }
+        ]
+    },
+    {
+        "id": "b304507",
+        "images": [
+            "b304507_00.jpg",
+            "b304507_01.jpg",
+            "b304507_02.jpg",
+            "b304507_03.jpg",
+            "b304507_04.jpg",
+            "b304507_05.jpg",
+            "b304507_06.jpg",
+            "b304507_07.jpg",
+            "b304507_08.jpg",
+            "b304507_09.jpg",
+            "b304507_10.jpg",
+            "b304507_11.jpg",
+            "b304507_12.jpg",
+            "b304507_13.jpg",
+            "b304507_14.jpg",
+            "b304507_15.jpg"
+        ],
+        "conversations": [
+            {
+                "from": "human",
+                "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is the parents are fighting, the child in the middle is crying  ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> \n"
+            },
+            {
+                "from": "gpt",
+                "value": "visual quality: 2\n object consistency: 3\n dynamic degree: 1\n motion smoothness: 1\n text-to-video alignment: 2\n factual consistency: 1\n overall score: 1\n"
+            }
+        ]
+    },
+    {
+        "id": "b005210",
+        "images": [
+            "b005210_00.jpg",
+            "b005210_01.jpg",
+            "b005210_02.jpg",
+            "b005210_03.jpg",
+            "b005210_04.jpg",
+            "b005210_05.jpg",
+            "b005210_06.jpg",
+            "b005210_07.jpg",
+            "b005210_08.jpg",
+            "b005210_09.jpg",
+            "b005210_10.jpg",
+            "b005210_11.jpg",
+            "b005210_12.jpg",
+            "b005210_13.jpg",
+            "b005210_14.jpg",
+            "b005210_15.jpg",
+            "b005210_16.jpg",
+            "b005210_17.jpg",
+            "b005210_18.jpg",
+            "b005210_19.jpg",
+            "b005210_20.jpg",
+            "b005210_21.jpg",
+            "b005210_22.jpg",
+            "b005210_23.jpg"
+        ],
+        "conversations": [
+            {
+                "from": "human",
+                "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is a golden king monkey, empire, cute colors, epic, cinematic  ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> \n"
+            },
+            {
+                "from": "gpt",
+                "value": "visual quality: 2\n object consistency: 1\n dynamic degree: 3\n motion smoothness: 1\n text-to-video alignment: 2\n factual consistency: 3\n overall score: 1\n"
+            }
+        ]
+    },
+    {
+        "id": "b303570",
+        "images": [
+            "b303570_00.jpg",
+            "b303570_01.jpg",
+            "b303570_02.jpg",
+            "b303570_03.jpg",
+            "b303570_04.jpg",
+            "b303570_05.jpg",
+            "b303570_06.jpg",
+            "b303570_07.jpg",
+            "b303570_08.jpg",
+            "b303570_09.jpg",
+            "b303570_10.jpg",
+            "b303570_11.jpg",
+            "b303570_12.jpg",
+            "b303570_13.jpg",
+            "b303570_14.jpg",
+            "b303570_15.jpg"
+        ],
+        "conversations": [
+            {
+                "from": "human",
+                "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is The camera slowly moves away from the cross as the sun sets on the horizon  ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> \n"
+            },
+            {
+                "from": "gpt",
+                "value": "visual quality: 3\n object consistency: 3\n dynamic degree: 1\n motion smoothness: 1\n text-to-video alignment: 1\n factual consistency: 1\n overall score: 1\n"
+            }
+        ]
+    },
+    {
+        "id": "b304106",
+        "images": [
+            "b304106_00.jpg",
+            "b304106_01.jpg",
+            "b304106_02.jpg",
+            "b304106_03.jpg",
+            "b304106_04.jpg",
+            "b304106_05.jpg",
+            "b304106_06.jpg",
+            "b304106_07.jpg",
+            "b304106_08.jpg",
+            "b304106_09.jpg",
+            "b304106_10.jpg",
+            "b304106_11.jpg",
+            "b304106_12.jpg",
+            "b304106_13.jpg",
+            "b304106_14.jpg",
+            "b304106_15.jpg"
+        ],
+        "conversations": [
+            {
+                "from": "human",
+                "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is cinematic vibrant action scene, two men in a chase crash through the window of the building, inception, matrix looking character HD, 4k resolution, sharp, red dragon cinema camera  ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> \n"
+            },
+            {
+                "from": "gpt",
+                "value": "visual quality: 3\n object consistency: 1\n dynamic degree: 3\n motion smoothness: 2\n text-to-video alignment: 1\n factual consistency: 2\n overall score: 1\n"
+            }
+        ]
+    }
+]