Motion Capture from Internet Videos
Junting Dong*, Qing Shuai*, Yuanqing Zhang, Xian Liu, Xiaowei Zhou, Hujun Bao
ECCV 2020 Project Page
You can download our modified Human3.6M dataset here.
First, we split the origin videos into different folders, and store the 3D annotations as follows.
<path_to_data>
├── data_2d_h36m_cpn_ft_h36m_dbb.npz
├── joints3d
│ ├── S9_Directions 1.mat
│ ├── S9_Directions.mat
│ ├── ...
│ ├── ...
│ ├── ...
│ ├── S9_WalkTogether 1.mat
│ └── S9_WalkTogether.mat
└── S9
├── Directions
│ ├── Directions.54138969.mp4
│ ├── Directions.55011271.mp4
│ ├── Directions.58860488.mp4
│ └── Directions.60457274.mp4
├── Directions1
│ ├── Directions1.54138969.mp4
│ ├── Directions1.55011271.mp4
│ ├── Directions1.58860488.mp4
│ └── Directions1.60457274.mp4
| ......
├── WalkTogether
│ ├── WalkTogether.54138969.mp4
│ ├── WalkTogether.55011271.mp4
│ ├── WalkTogether.58860488.mp4
│ └── WalkTogether.60457274.mp4
└── WalkTogether1
├── ......
We use finetune cpn output as our 2D pose from videopose3d
wget https://dl.fbaipublicfiles.com/video-pose-3d/data_2d_h36m_cpn_ft_h36m_dbb.npzAfter all, you can generate the synthetic data. More details can be found in the file script/dataset/sample_h36m.py.
python3 script/dataset/sample_h36m.py --video_path <path_to_data>/S9Our quantitative evaluation includes two parts: match and reconstruction. We provide the evaluation scripts as example.