This repository is the implementation code and model of the paper "HandTailor: Towards High-Precision Monocular 3D Hand Recovery" (BMVC2021) (paper). You can find the video demo on the webpage.

git clone https://github.com/LyuJ1998/HandTailor.git
cd HandTailor

Please install the dependencies listed in requirements.txt.

pip install -r requirements.txt

Download the pretrain model from Google Drive, and put the model.pt in ./checkpoints

• Go to MANO website
• Create an account by clicking Sign Up and provide your information
• Download Models and Code (the downloaded file should have the format mano_v*_*.zip). Note that all code and data from this download falls under the MANO license.
• unzip and copy the MANO_RIGHT.pkl file into the folder

To process the image provided in ./demo, run

python demo.py

You can also put your data in the fold ./demo, but remember to use the proper camera intrinsic like

python demo.py --fx=612.0206 --fy=612.2821 --cx=321.2842 --cy=235.8609

If camera information is unavailable, run

python demo_in_the_wild.py

We recommand you to utilize the camera intrinsic, which will improve the performance a lot.

To reconstruct the hand from image captured with a webcam,run the following command. Also remember to use the proper camera intrinsic, the following command is for RealSense D435

python app.py --fx=612.0206 --fy=612.2821 --cx=321.2842 --cy=235.8609

When camera information is absence

python app_in_the_wild.py

When using HandTailor to recovery hand mesh, you need to make sure that the hand is in the dominate area of the image. To address this, we also implement a naive tracker.

Please run:

python app_with_tracker.py --fx=612.0206 --fy=612.2821 --cx=321.2842 --cy=235.8609

This is a quite simple tracker, so do not move your hand too fast. And once track lost, put your hand on the bounding box to fix it.

If you find this work helpful, please consider citing us

@article{lv2021handtailor,
  title     = {HandTailor: Towards High-Precision Monocular 3D Hand Recovery},
  author    = {Lv, Jun and Xu, Wenqiang and Yang, Lixin and Qian, Sucheng and Mao, Chongzhao and Lu, Cewu},
  booktitle = {British Machine Vision Conference (BMVC)},
  year      = {2021}
}