This repository contains PyTorch implementation of the NeurIPS paper:

OST: Improving Generalization of DeepFake Detection via One-Shot Test-Time Training.

Liang Chen, Yong Zhang, Yibing Song, Jue Wang, Lingqiao Liu

A new learning paradigm specially designed for the deepfake detection task:

Preparation

Same as that in SLADD:

Download Xception pretrained weights and dlib landmark predictor and put them in the weights folder.

If you find this code useful for your research, please cite:

@inproceedings{chen2022ost,
title={OST: Improving generalization of deepfake detection via one-shot test-time training},
author={Chen, Liang and Zhang, Yong and Song, Yibing and Wang, Jue and Liu, Lingqiao},
booktitle={Advances in Neural Information Processing Systems},
year={2022}}

Please emial me directly ([email protected]) if you have any questions or any feedback (since I no longer work on the deepfake detection task, issues in this project may not be properly resolved in time).