Patch-Discontinuity Mining for Generalized Deepfake Detection

TL;DR

This paper introduces GenDF, a lightweight deepfake detection framework that leverages large-scale vision models and novel representation learning techniques to improve cross-domain generalization.

Contribution

The paper proposes a simple, parameter-efficient framework that enhances deepfake detection generalization using representation learning, feature redistribution, and invariant feature augmentation.

Findings

Achieves state-of-the-art cross-domain detection performance.

Requires only 0.28 million trainable parameters.

Demonstrates strong generalization across manipulation methods.

Abstract

The rapid advancement of generative artificial intelligence has enabled the creation of highly realistic fake facial images, posing serious threats to personal privacy and the integrity of online information. Existing deepfake detection methods often rely on handcrafted forensic cues and complex architectures, achieving strong performance in intra-domain settings but suffering significant degradation when confronted with unseen forgery patterns. In this paper, we propose GenDF, a simple yet effective framework that transfers a powerful large-scale vision model to the deepfake detection task with a compact and neat network design. GenDF incorporates deepfake-specific representation learning to capture discriminative patterns between real and fake facial images, feature space redistribution to mitigate distribution mismatch, and a classification-invariant feature augmentation strategy to enhance generalization without introducing additional trainable parameters. Extensive experiments demonstrate that GenDF achieves state-of-the-art generalization performance in cross-domain and cross-manipulation settings while requiring only 0.28M trainable parameters, validating the effectiveness and efficiency of the proposed framework.