For easy installation, we provide all set-up docker files. File Size (36.6GB) We recommend using Docker file. Docker File Link

pip install -r requirements.txt

pip install -r optional-requirements.txt

Follow the installation instructions for each of the following before using the fitting code.

1. PyTorch
2. SMPL-X
3. VPoser
4. Homogenus

Generating hand mesh is based on smplx, download the model go to this project website and register to get access to the downloads section. and put the files to /model/smplx

1. PyTorch Mesh self-intersection for interpenetration penalty

2. Trimesh for loading triangular meshes

3. Pyrender for visualization

Openpose keypoint json file , whole body images

Move required files to the /data/images , /data/keypoints

/data directory contain two subfolders, images, where the images are located, and keypoints, where the OpenPose output should be stored.

./run.sh

This script will generate hand mesh iteratively and the ouput of the images will be stored in /data/output

The format of the output file contains an image with a mesh overlapping. and .obj file

You can change the max iteration value by changing the yaml file in /cfg_files/fit_smplx.yaml.

#Max number of optimization iterations
maxiters: 100

Change the value of the maxiters in the yaml file

https://github.com/kuai-lab/iterative_fitting_hand/blob/c7617c0748f9cb59669d6dd3d52c4bce7cfc00e7/cfg_files/fit_smplx.yaml#L44

You can visualize the hand mesh currently being generated by image. Uncomment the code from smplifyx/fit_single_frame.py

https://github.com/kuai-lab/iterative_fitting_hand/blob/c7617c0748f9cb59669d6dd3d52c4bce7cfc00e7/smplifyx/fit_single_frame.py#L539

#hand_mesh.show()

https://github.com/kuai-lab/iterative_fitting_hand/blob/c7617c0748f9cb59669d6dd3d52c4bce7cfc00e7/smplifyx/fit_single_frame.py#L544

#scene.show()