Introduction

The official PyTorch implementation for the following paper:

Fully Unsupervised Deepfake Video Detection via Enhanced Contrastive Learning

Paper link: (https://ieeexplore.ieee.org/abstract/document/10411047)

Environment

Python==3.7.6, torch==1.10.1, torchvision==0.11.2, cudatoolkit==11.1

Run

Stage 1 code: pseudo_label_generator.py

Stage 2 code: enhanced_contrastive_learner.py

Stage 3 code: In release

Citations

Please cite the following paper in your publications if you use the python implementations:

@article{qiao2024fully,
  title={Fully Unsupervised Deepfake Video Detection via Enhanced Contrastive Learning},
  author={Qiao, Tong and Xie, Shichuang and Chen, Yanli and Retraint, Florent and Luo, Xiangyang},
  journal={IEEE Transactions on Pattern Analysis and Machine Intelligence},
  year={2024},
  publisher={IEEE}
}

Acknowledgments

SimCLR.

SupContrast

EVA