retarget-toolbox

Motion Retargeting from Human Motion to Humanoid Robots Based on Gradient Descent with a User-Friendly Interface

🌟 Key Features

• Batch Processing: Efficiently process multiple human mocap files in one operation.
• Motion Visualization: Visualize motion trajectories and keyframes.

📦 Installation

conda create -n ret python=3.9
conda activate ret
conda install pytorch==2.2.0 torchvision==0.17.0 torchaudio==2.2.0 pytorch-cuda=11.8 -c pytorch -c nvidia
conda install pyg -c pyg
pip install urdfpy matplotlib h5py pandas scikit-learn
conda install -c conda-forge networkx=2.5
pip install "numpy<2"
# Install visualization module
cd viz
pip install -e .

🚀 Get Started

Data Preprocessing

# Requires BlenderPy: https://github.com/TylerGubala/blenderpy
# Otherwise, use the example motion files in the `data` folder
cd preprocess
python mirror_bvh.py
python transform_bvh.py

Whole-body Motion Retargeting

cd retarget
python optimize_hi.py

Visualization

cd viz
# Human control with sliders
python example/human_control.py
# Play H5 File
python example/h5_control.py

📄 Citation

If you use this project in your research, please cite it as follows:

@misc{retarget-toolbox,
author = {Haodong Zhang},
title = {Human-Robot Motion Retargeting Toolbox},
year = {2025},
publisher = {GitHub},
howpublished = {\url{https://github.com/0aqz0/retarget-toolbox}},
}