Towards Generalizable Deepfake Detection by Primary Region Regularization

TL;DR

This paper proposes a novel primary region regularization method to improve deepfake detection generalization by reducing overfitting to specific regions, enhancing robustness across unseen forgeries.

Contribution

It introduces a primary region removal augmentation technique and a two-stage localization and exploitation framework that can be integrated into various models.

Findings

Achieves an average 6% performance boost across multiple datasets.

Improves generalization to unseen deepfake forgeries.

Performs competitively with state-of-the-art methods.

Abstract

The existing deepfake detection methods have reached a bottleneck in generalizing to unseen forgeries and manipulation approaches. Based on the observation that the deepfake detectors exhibit a preference for overfitting the specific primary regions in input, this paper enhances the generalization capability from a novel regularization perspective. This can be simply achieved by augmenting the images through primary region removal, thereby preventing the detector from over-relying on data bias. Our method consists of two stages, namely the static localization for primary region maps, as well as the dynamic exploitation of primary region masks. The proposed method can be seamlessly integrated into different backbones without affecting their inference efficiency. We conduct extensive experiments over three widely used deepfake datasets - DFDC, DF-1.0, and Celeb-DF with five backbones. Our method demonstrates an average performance improvement of 6% across different backbones and performs competitively with several state-of-the-art baselines.