🚀 Track video 🚀

🚅 Track image 🚅

GAGAvatar Track is a monocular face tracker built on FLAME. It provides FLAME parameters (including eyeball pose) and camera parameters, along with the bounding box and landmarks used during optimization.

This environment is a sub-environment of GAGAvatar. You can skip this step if you have already built GAGAvatar.

conda env create -f environment.yml
conda activate GAGAvatar_track

Prepare resources with bash ./build_resources.sh.

It takes longer to track the first frame.

python track_video.py -v ./demos/obama.mp4

python track_image.py -i ./demos/monroe.jpg

python track_lmdb.py -l ./demos/vfhq_demo

If you find our work useful in your research, please consider citing:

@inproceedings{
    chu2024gagavatar,
    title={Generalizable and Animatable Gaussian Head Avatar},
    author={Xuangeng Chu and Tatsuya Harada},
    booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems},
    year={2024},
    url={https://openreview.net/forum?id=gVM2AZ5xA6}
}

Some part of our work is built based on FLAME, StyleMatte, EMICA and VGGHead. The GAGAvatar Logo is designed by Caihong Ning. We thank you for sharing their wonderful code and their wonderful work.

• FLAME: https://flame.is.tue.mpg.de
• StyleMatte: https://github.com/chroneus/stylematte
• EMICA: https://github.com/radekd91/inferno
• VGGHead: https://github.com/KupynOrest/head_detector