Skip to content

GANG370/FedForgery

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

7 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

FedForgery: Generalized Face Forgery Detection with Residual Federated Learning

The zip file contains the source code we used in this paper to test the accuracy of face forgery detection in the Hybrid-domain forgery dataset.

Dependencies

  • Anaconda3 (Python3.6, with Numpy etc.)
  • Pytorch 1.10.0
  • tensorboardX

More details about dependencies are shown in requirements.txt.

Datasets

[Hybrid-domain forgery dataset] Combine four diverse forgery subtypes of the FF++ dataset and the WildDeepfake dataset into the whole dataset with five different artifact types the training set contains 20,000 images where true images have the same number of fake images; the ratio of the training set and testing set is kept at 7: 3.]

Usage

Download pretrained model

Model Download
FedForgery MEGA

After downloading the pretrained model, we should put the model to ./pretrained

Download dataset

Dataset Name Download Images
Hybrid-domain forgery dataset Hybrid-domain forgery- dataset 4,2800

After downloading the whole dataset, you can unzip test.zip to the ./testset.

Test the model

./run_test.sh

BibTeX


@article{liu2023fedforgery,
  title={FedForgery: generalized face forgery detection with residual federated learning},
  author={Liu, Decheng and Dang, Zhan and Peng, Chunlei and Zheng, Yu and Li, Shuang and Wang, Nannan and Gao, Xinbo},
  journal={IEEE Transactions on Information Forensics and Security},
  year={2023},
  publisher={IEEE}
}

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published