CoMoVi: Co-Generation of 3D Human Motions
and Realistic Videos

Chengfeng Zhao¹, Jiazhi Shu², Yubo Zhao¹, Tianyu Huang³, Jiahao Lu¹,
Zekai Gu¹, Chengwei Ren¹, Zhiyang Dou⁴, Qing Shuai⁵, Yuan Liu¹

¹HKUST    ²SCUT    ³CUHK    ⁴MIT    ⁵ZJU   
Corresponding author

🚀 Getting Started

1. Environment Setup

conda create python=3.10 --name comovi
conda activate comovi

pip install torch==2.5.0 torchvision==0.20.0 torchaudio==2.5.0 --index-url https://download.pytorch.org/whl/cu121
pip install -r requirements.txt

pip install ninja
pip install flash_attn --no-build-isolation # ==2.7.3 for CUDA < 12

pip install git+https://github.com/facebookresearch/detectron2

conda install -c fvcore -c iopath -c conda-forge fvcore iopath
pip install "git+https://github.com/facebookresearch/pytorch3d.git@stable"

2. Inference

bash scripts/inference.sh

Explanation of arguments:

• arch
• validation_file: to be deleted
• exp_name: to be merged with ckpt_at
• fps: frame rate of generated video, default is 16
• frames: frame num of generated video, default is 81
• height: H of generated video, default is 704
• width: W of generated video, default is 1280
• ckpt_at: to be merged with exp_name
• motion_type: to be deleted
• interaction: to be deleted
• interleave: to be deleted
• nodebug: to be deleted

🔬 Training

1. Data Preparation

Install Blender

mkdir <dir_for_blender>
cd <dir_for_blender>

wget https://download.blender.org/release/Blender3.6/blender-3.6.0-linux-x64.tar.xz
xz -d blender-3.6.0-linux-x64.tar.xz
tar -xvf blender-3.6.0-linux-x64.tar

export PATH=<dir_for_blender>/blender-3.6.0-linux-x64:$PATH

Install CameraHMR

bash scripts/install_camerahmr.sh

Option-1: Download CoMoVi dataset

Coming soon.

Option-2: Pepare customized data step by step

Step-1: Estimate human motion from image frames

python -m prepare.step1_run_hmr  

Step-2: Smooth framewise motion estimation

python -m prepare.step2_smooth

Step-3: Render 3D human motion to 2D motion representation

python -m prepare.step3_render_2d_morep

After the three steps above, your examples/ folder should have the following structure:

examples/
├── CameraHMR_smpl_results/           # raw HMR results
└── CameraHMR_smpl_results_overlay/   # raw HMR re-projection results for sanity check
└── CameraHMR_smpl_results_smoothed/  # smoothed HMR results
└── motion_2d_videos/                 # rendered 2d motion representation video
└── rgb_videos/                       # rgb video 

Step-4: Normalize data to the native setting of Wan2.2 (e.g. resolution, fps, etc.)

python -m prepare.step4_normalize

Step-5: Caption description of human motion in videos

2. Train CoMoVi

bash scripts/wan2.2/train_puma_multinode_motion_branch_add_smpl.sh

Acknowledgments

Thanks to the following work that we refer to and benefit from:

• VideoX-Fun: the video generation model training framework;
• CameraHMR: the excellent SMPL estimation for pseudo labels;
• Champ: the data processing pipeline

Citation

@article{zhao2026comovi,
  title={CoMoVi: Co-Generation of 3D Human Motions and Realistic Videos},
  author={Zhao, Chengfeng and Shu, Jiazhi and Zhao, Yubo and Huang, Tianyu and Lu, Jiahao and Gu, Zekai and Ren, Chengwei and Dou, Zhiyang and Shuai, Qing and Liu, Yuan},
  journal={arXiv preprint arXiv:2601.10632},
  year={2026}
}