Official implementation of the CVPR 2024 paper "FSRT: Facial Scene Representation Transformer for Face Reenactment from Factorized Appearance, Head-pose, and Facial Expression Features"
until the reqs are uploaded i did this to get it working
as im using cuda 11.8 i did this for the pytorch
conda install pytorch torchvision torchaudio pytorch-cuda=11.8 -c pytorch -c nvidia
until the reqs are uploaded i did this to get it working as im using cuda 11.8 i did this for the pytorch
conda install pytorch torchvision torchaudio pytorch-cuda=11.8 -c pytorch -c nvidia
then the other reqs are as follows
pip install tqdm
pip install matplotlib
pip install einops
pip install ffmpeg
pip install imageio[ffmpeg]
pip install imageio[pyav]
for the checkpoints i put the kp_detector in a new folder called weights ./weights/kp_detector.pt and ./fsrt_checkpoints/vox256.pt
I also create a folder called examples to keep the driving and source in ill provide them here
https://drive.google.com/drive/folders/1a0lk0NUqwDj8-2OzDgGcGU2cOGRuSGAT?usp=sharing