FakePolisher: Making DeepFakes More Detection-Evasive by Shallow Reconstruction

TL;DR

FakePolisher is a shallow reconstruction method that reduces artifacts in GAN-generated images, making DeepFake detection methods less effective and highlighting challenges in current detection techniques.

Contribution

The paper introduces FakePolisher, a novel shallow reconstruction approach that effectively reduces artifacts in GAN images to evade detection.

Findings

Reduces detection accuracy by 47% on average

Effective across 16 GAN-based generation techniques

Significantly diminishes artifact patterns in fake images

Abstract

At this moment, GAN-based image generation methods are still imperfect, whose upsampling design has limitations in leaving some certain artifact patterns in the synthesized image. Such artifact patterns can be easily exploited (by recent methods) for difference detection of real and GAN-synthesized images. However, the existing detection methods put much emphasis on the artifact patterns, which can become futile if such artifact patterns were reduced. Towards reducing the artifacts in the synthesized images, in this paper, we devise a simple yet powerful approach termed FakePolisher that performs shallow reconstruction of fake images through a learned linear dictionary, intending to effectively and efficiently reduce the artifacts introduced during image synthesis. The comprehensive evaluation on 3 state-of-the-art DeepFake detection methods and fake images generated by 16 popular GAN-based fake image generation techniques, demonstrates the effectiveness of our technique.Overall, through reducing artifact patterns, our technique significantly reduces the accuracy of the 3 state-of-the-art fake image detection methods, i.e., 47% on average and up to 93% in the worst case.