Enhancing Abnormality Identification: Robust Out-of-Distribution Strategies for Deepfake Detection

2025-06-03

AI Summary

Existing deepfake detection methods operate under a closed-world assumption, leading to severe performance degradation in open-world scenarios when encountering unseen generative models. Method: This paper proposes a dual-path out-of-distribution (OOD) detection framework that jointly leverages self-supervised image reconstruction for global inconsistency modeling and attention-driven feature disentanglement for fine-grained anomaly localization; it further incorporates OOD confidence calibration to enhance decision robustness. Contributions/Results: The framework achieves state-of-the-art performance across multiple mainstream benchmarks. It significantly improves detection accuracy on deepfakes generated by previously unseen models, effectively mitigating the limitations imposed by the closed-world assumption and advancing generalizable, open-set deepfake detection.

Technology Category

• Computer Vision: Adversarial Attacks & Robustness
• Machine Learning: Calibration & Uncertainty Quantification
• Data Mining & Knowledge Management: Anomaly/Outlier Detection

Application Category

• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Detecting deepfakes has become a critical challenge in Computer Vision and Artificial Intelligence. Despite significant progress in detection techniques, generalizing them to open-set scenarios continues to be a persistent difficulty. Neural networks are often trained on the closed-world assumption, but with new generative models constantly evolving, it is inevitable to encounter data generated by models that are not part of the training distribution. To address these challenges, in this paper, we propose two novel Out-Of-Distribution (OOD) detection approaches. The first approach is trained to reconstruct the input image, while the second incorporates an attention mechanism for detecting OODs. Our experiments validate the effectiveness of the proposed approaches compared to existing state-of-the-art techniques. Our method achieves promising results in deepfake detection and ranks among the top-performing configurations on the benchmark, demonstrating their potential for robust, adaptable solutions in dynamic, real-world applications.

Problem

• Detecting deepfakes in open-set scenarios
• Improving generalization of deepfake detection techniques
• Addressing data from evolving generative models

Innovation

• Reconstruct input image for OOD detection
• Use attention mechanism to detect OODs
• Robust deepfake detection in dynamic scenarios