FrePGAN: Robust Deepfake Detection Using Frequency-level Perturbations

TL;DR

This paper introduces FrePGAN, a deepfake detection framework that leverages frequency-level perturbations to improve detection of both known and unknown GAN-generated images, achieving state-of-the-art results.

Contribution

The paper proposes a novel frequency-level perturbation approach combined with a training framework to enhance deepfake detector generalization across unseen GAN models.

Findings

Improved detection accuracy on unseen GAN models.

Enhanced robustness against various image manipulations.

State-of-the-art performance in diverse test scenarios.

Abstract

Various deepfake detectors have been proposed, but challenges still exist to detect images of unknown categories or GAN models outside of the training settings. Such issues arise from the overfitting issue, which we discover from our own analysis and the previous studies to originate from the frequency-level artifacts in generated images. We find that ignoring the frequency-level artifacts can improve the detector's generalization across various GAN models, but it can reduce the model's performance for the trained GAN models. Thus, we design a framework to generalize the deepfake detector for both the known and unseen GAN models. Our framework generates the frequency-level perturbation maps to make the generated images indistinguishable from the real images. By updating the deepfake detector along with the training of the perturbation generator, our model is trained to detect the frequency-level artifacts at the initial iterations and consider the image-level irregularities at the last iterations. For experiments, we design new test scenarios varying from the training settings in GAN models, color manipulations, and object categories. Numerous experiments validate the state-of-the-art performance of our deepfake detector.