pose estimation for PD
- Environment
- Linux system
- Python > 3.6 distribution
- Dependencies
- Packages
- Pytorch > 1.0.0
- torchsample
- ffmpeg
- tqdm
- pillow
- scipy
- pandas
- h5py
- visdom
- nibabel
- opencv-python (install with pip)
- matplotlib
- Packages
- place your video into
./input
folder. (There are thre test videos there).
- change the
video_path
in the./videopose_PSTMO.py
- Run it! You will find the rendered output video in the
./outputs
folder.
The codes are based on P-STMO codes, see https://github.com/paTRICK-swk/P-STMO
The 2D pose to 3D pose and visualization part is from VideoPose3D.
Some of the "In the wild" script is adapted from the other fork.
The project structure and ./videopose.py
running script is adapted from this repo
The other feature will be added to improve accuracy in the future:
- more data to be tested
- Use PD data as our training data
- use the 3D joints data to drive the neuromuscular model
- Change 2D pose estimation method such as AlphaPose.
- Test HR-Net as 2d joints detector.
- Test LightTrack as pose tracker.