This is the official PyTorch implementation of MoCaNet: Motion Retargeting in-the-wild via Canonicalization Networks (AAAI 2022).
Paper | Project Page | Poster
conda install pytorch torchvision cudatoolkit=10.1 -c pytorch
conda install pyyaml scikit-image scikit-learn opencv
pip install -r requirements.txt
We use the Mixamo and SoloDancer datasets. Please refer to TransMoMo for detailed instructions.
for descriptions of training options, do
python train.py --help
To train on Mixamo
python train.py --config configs/mocanet.yaml
To train on SoloDancer
python train.py --config configs/mocanet_solodancer.yaml
If you use our code or models in your research, please cite with:
W. Zhu*, Z. Yang*, Z. Di, W. Wu+, Y. Wang, C. C. Loy. "MoCaNet: Motion Retargeting in-the-wild via Canonicalization Networks." Thirty-Sixth AAAI Conference on Artificial Intelligence (AAAI), 2022.
BibTeX:
@inproceedings{mocanet2022,
title={MoCaNet: Motion Retargeting in-the-wild via Canonicalization Networks},
author={Zhu, Wentao and Yang, Zhuoqian and Di, Ziang and Wu, Wayne and Wang, Yizhou and Loy, Chen Change},
booktitle={AAAI},
year={2022}
}