The official code for CMUA-Watermark: A Cross-Model Universal Adversarial Watermark for Combating Deepfakes (AAAI2022) arxiv. It is based on disrupting-deepfakes .
Contact us with [email protected], [email protected].
We will release our code soon (no later than December 31, 2021).
CMUA-Watermark is a cross-model universal adversarial watermark that can combat multiple deepfake models while protecting a myriad of facial images. With the proposed perturbation fusion strategies and automatic step size tuning, CMUA-Watermark achieves excellent protection capabilities for facial images against four face modification models (StarGAN, AttGAN, AGGAN, HiSD).
Figure 1. Illustration of our CMUA-Watermark. Once the CMUA-watermark has been generated, we can add it directly to any facial image to generate a protected image that is visually identical to the original image but can distort outputs of deepfake models.Figure 2. The quantitative results of CMUA-Watermark.
-
Prepare the Environment
-
(option1) install the lib by pip (recommend)
pip install -r requirements.txt
-
(option2) We also prepare conda environment for you (if you use CUDA 10.0 and conda 4.8.3) : https://disk.pku.edu.cn:443/link/D613E493EE641184EB52C0C78DD846C8 . You can donwload it and unzip in your '/anaconda3/envs/'.
-
-
Prepare the Datasets
- download the CelebA datasets:
make sure your floder (e.g. celeba_data) has img_align_celeba and list_attr_celeba.txt.
cd stargan bash download.sh celeba
- create the link
ln -s your_path_to_celeba_data ./data
- download the CelebA datasets:
-
Prepare the Model Weights
For your convenient usage, we prepare the weights download link in PKU disk: https://disk.pku.edu.cn:443/link/D04A3ED9D22694D81924109D0E4EACA8.
You can first download the weights. Then move the weight files to different floders of different models:
cd CMUA-Watermark # make sure **weights** in this path. # If the paths bellow are not exist, please create the path (e.g., mkdir -p ./stargan/stargan_celeba_256/models). mv ./weights/stargan/* ./stargan/stargan_celeba_256/models mv ./weights/AttentionGAN/* ./AttentionGAN/AttentionGAN_v1_multi/checkpoints/celeba_256_pretrained mv ./weights/HiSD/* ./HiSD mv ./weights/AttGAN/* ./AttGAN/output/256_shortcut1_inject0_none/checkpoint
ATTENTION! The copyright of these weight files belongs to their owners. You needs authorization for commerce, please contact to their owners!
-
Prepare the CMUA-Watermark (only for inference)
We prepare a CMUA-Watermark for you to test its performance: https://disk.pku.edu.cn:443/link/4FDBB772471746EC0DC397B520005D3E.
# inference in CelebA datasets with 20 images (you can change the test number in evaluate.py)
python3 universal_attack_inference.py
# inference with your own image (one image)
python3 universal_attack_inference_one_image.py ./demo_input.png # you can change the path with your own image
If your want to try your onw idea, you may need to modify the nni_config.yaml and search_space.json. These two files are the configs of NNI-based search. Thanks to the NNI, you can obtain the visualized results in your browser.
nnictl create --config ./nni_config.yaml
Once you get the best step sizes, you need to modify the default step sizes in setting.json. It must be easy for a smart person like you~
After that,
python universal_attack.py
If you use our code / perturbation, please consider to cite our paper: CMUA-Watermark: A Cross-Model Universal Adversarial Watermark for Combating Deepfakes.
@misc{huang2021cmuawatermark,
title={CMUA-Watermark: A Cross-Model Universal Adversarial Watermark for Combating Deepfakes},
author={Hao Huang and Yongtao Wang and Zhaoyu Chen and Yuze Zhang and Yuheng Li and Zhi Tang and Wei Chu and Jingdong Chen and Weisi Lin and Kai-Kuang Ma},
year={2021},
eprint={2105.10872},
archivePrefix={arXiv},
primaryClass={cs.CV}
}
The project is only free for academic research purposes, but needs authorization for commerce. For commerce permission, please contact [email protected].
We use code from StarGAN, GANimation, pix2pixHD, CycleGAN, advertorch, disrupting-deepfakes and nni. These are all great repositories and we encourage you to check them out and cite them in your work.