VTON-HandFit: Virtual Try-on for Arbitrary Hand Pose Guided by Hand Priors Embedding

This is the offical code repository of “VTON-HandFit: Virtual Try-on for Arbitrary Hand Pose Guided by Hand Priors Embedding.”

Installation

We recommend creating a virtual environment for Handfit, you can create it with

conda env create -f environment.yml
conda activate handfit
pip install torch==2.4.0 torchvision==0.19.0 xformers==0.0.27.post2 --index-url https://download.pytorch.org/whl/cu118

Install hamer

cd hamer
pip install -e .[all]
pip install pyopengl==3.1.4

Install pytorch3d

pip install git+https://github.com/facebookresearch/pytorch3d.git@stable

Install DensePose

pip install git+https://github.com/facebookresearch/detectron2@main#subdirectory=projects/DensePose

Install ViTPose

cd third-party/ViTPose/
pip install -v -e .

Citation

@article{liang2024vton,
  title={VTON-HandFit: Virtual Try-on for Arbitrary Hand Pose Guided by Hand Priors Embedding},
  author={Liang, Yujie and Hu, Xiaobin and Jiang, Boyuan and Luo, Donghao and Wu, Kai and Han, Wenhui and Jin, Taisong and Wang, Chengjie},
  journal={arXiv preprint arXiv:2408.12340},
  year={2024}
}

License

The codes and checkpoints in this repository are under the CC BY-NC-SA 4.0 license.