Towards Universal GAN Image Detection

TL;DR

This paper introduces a new GAN image detection method that uses sub-sampling and contrastive learning, demonstrating improved robustness and generalization to unseen GAN architectures in challenging real-world conditions.

Contribution

It proposes a novel detection approach combining sub-sampling and contrastive learning, addressing robustness and universality issues of prior GAN detectors.

Findings

Effective in challenging conditions

Robust to common image impairments

Generalizes well to unseen architectures

Abstract

The ever higher quality and wide diffusion of fake images have spawn a quest for reliable forensic tools. Many GAN image detectors have been proposed, recently. In real world scenarios, however, most of them show limited robustness and generalization ability. Moreover, they often rely on side information not available at test time, that is, they are not universal. We investigate these problems and propose a new GAN image detector based on a limited sub-sampling architecture and a suitable contrastive learning paradigm. Experiments carried out in challenging conditions prove the proposed method to be a first step towards universal GAN image detection, ensuring also good robustness to common image impairments, and good generalization to unseen architectures.