shashwatsrii/Minor-Project

Adversarial Filter for Deepfake Prevention

Minor-Project

Adversarial Filter for Deepfake Prevention

@article{wang2022anti, title={Anti-Forgery: Towards a Stealthy and Robust DeepFake Disruption Attack via Adversarial Perceptual-aware Perturbations}, author={Wang, Run and Huang, Ziheng and Chen, Zhikai and Liu, Li and Chen, Jing and Wang, Lina}, journal={arXiv preprint arXiv:2206.00477}, year={2022} }

Step 1 : git clone https://github.com/shashwatsrii/Minor-Project/tree/main

For Lab Color Space

Step 2 : cd LabColorSpace

Step 3 : pip install -r requirements.txt

Step 4 : icacls download.sh /grant Everyone:RX

Step 5 : bash download.sh celeba

Step 6 : python main.py --mode test --image_size 256 --c_dim 5 --selected_attrs Black_Hair Blond_Hair Brown_Hair Male Young --model_save_dir='stargan_celeba_256/models' --result_dir='./results' --test_iters 200000 --attack_iters 100 --batch_size 1