Advancing Generalized Deepfake Detector with Forgery Perception Guidance

Ruiyang Xia, Dawei Zhou, Decheng Liu, Lin Yuan, Shuodi Wang, Jie Li, Nannan Wang, Xinbo Gao

Abstract: One of the serious impacts brought by artificial intelligence is the abuse of deepfake techniques. Despite the proliferation of deepfake detection methods aimed at safeguarding the authenticity of media across the Internet, they mainly consider the improvement of detector architecture or the synthesis of forgery samples. The forgery perceptions, including the feature responses and prediction scores for forgery samples, have not been well considered. As a result, the generalization across multiple deepfake techniques always comes with complicated detector structures and expensive training costs. In this paper, we shift the focus to real-time perception analysis in the training process and generalize deepfake detectors through an efficient method dubbed Forgery Perception Guidance (FPG). In particular, after investigating the deficiencies of forgery perceptions, FPG adopts a sample refinement strategy to pertinently train the detector, thereby elevating the generalization efficiently. Moreover, FPG introduces more sample information as explicit optimizations, which makes the detector further adapt the sample diversities. Experiments demonstrate that FPG improves the generality of deepfake detectors with small training costs, minor detector modifications, and the acquirement of real data only. In particular, our approach not only outperforms the state-of-the-art on both the cross-dataset and cross-manipulation evaluation but also surpasses the baseline that needs more than 3$\times$ training time. Code is available in the supplementary material.

Primary Subject Area: [Generation] Social Aspects of Generative AI

Secondary Subject Area: [Content] Media Interpretation

Relevance To Conference: The development of artificial intelligence not only elevates performance in traditional visual tasks but also gives birth to massive novel and heuristic vision applications. Deepfake, a novel technique used to generate believable media via deep neural networks, has quickly developed and aroused social concerns due to the lifelikeness of the generation and the simplicity of usage. To ensure the safety and credibility of media oriented for the public, in the field of computer vision and multimedia research, deepfake detection methods have recently been proposed to discern the authenticity of media automatically.

Submission Number: 1620