Deepfake Forensics via An Adversarial Game

TL;DR

This paper proposes an adversarial training approach for deepfake detection that enhances model generalization to unseen forgeries and image qualities by using adversarially crafted samples and artifact-blurring techniques.

Contribution

It introduces a novel adversarial training method combining adversarial sample generation and artifact blurring to improve deepfake detection robustness.

Findings

Enhanced generalization to unseen forgeries

Improved robustness against image quality variations

Empirical validation shows significant performance gains

Abstract

With the progress in AI-based facial forgery (i.e., deepfake), people are increasingly concerned about its abuse. Albeit effort has been made for training classification (also known as deepfake detection) models to recognize such forgeries, existing models suffer from poor generalization to unseen forgery technologies and high sensitivity to changes in image/video quality. In this paper, we advocate adversarial training for improving the generalization ability to both unseen facial forgeries and unseen image/video qualities. We believe training with samples that are adversarially crafted to attack the classification models improves the generalization ability considerably. Considering that AI-based face manipulation often leads to high-frequency artifacts that can be easily spotted by models yet difficult to generalize, we further propose a new adversarial training method that attempts to blur out these specific artifacts, by introducing pixel-wise Gaussian blurring models. With adversarial training, the classification models are forced to learn more discriminative and generalizable features, and the effectiveness of our method can be verified by plenty of empirical evidence. Our code will be made publicly available.