Official PyTorch implementation of Diffusion-based Face Swapping with Facial Guidance
git clone https://github.com/hxngiee/DiffFace.git
cd DiffFace
conda create -n DiffFace python=3.9.7
conda activate DiffFace
conda install pytorch==1.10.0 torchvision==0.11.0 torchaudio==0.10.0 -c pytorch
pip install -r requirements.txt
The weights required for the execution of our DiffFace can be downloaded from link.
mkdir checkpoints
mv Arcface.tar checkpoints/
mv FaceParser.pth checkpoints/
mv GazeEstimator.pt checkpoints/
mv Model.pt checkpoints/
The dataset and checkpoints should be placed in the following structures below
DiffFace
├── checkpoints
├── Arcface.tar
├── FaceParser.pth
├── GazeEstimator.pt
└── Model.pt
├── data
└── src
├── 001.png
└── ...
└── targ
├── 001.png
└── ...
├── models
├── optimization
├── utils
└── main.py
Place source and target images in data/src, and data/targ. Then run the following.
python main.py --output_path output/example
This licence allows for academic and non-commercial purpose only. The entire project is under the CC-BY-NC 4.0 license.
If you find DiffFace useful for your work please cite:
@inproceedings{kim2022diffface,
title = {DiffFace: Diffusion-based Face Swapping with Facial Guidance},
author = {K. Kim, Y. Kim, S. Cho, J. Seo, J. Nam, K. Lee, S. Kim, K. Lee},
journal = {Arxiv},
year = {2022}
}
This code borrows heavily from Blended-Diffusion and Guided-Diffusion.
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent