DiffFake: Exposing Deepfakes using Differential Anomaly Detection

2025-02-22

AI Summary

Traditional deepfake detectors rely on binary classification, exhibiting poor generalization to unseen generation techniques. This work proposes a differential anomaly detection framework designed for cross-technique generalization: it reframes deepfake detection as identifying anomalies in natural variation patterns between two face images of the same identity, abandoning supervised binary classification. Key contributions include: (i) the first introduction of the differential anomaly detection paradigm for deepfake detection; (ii) a pseudo-deepfake–driven feature extractor that jointly models global and local forgery cues; and (iii) an integrated pipeline combining deep face embeddings, differential feature learning, self-supervised pseudo-sample generation, and deep anomaly detection (e.g., Deep SVDD). Evaluated on five mainstream benchmarks, the method achieves or surpasses state-of-the-art performance, significantly improves detection accuracy on unknown generation techniques, and demonstrates markedly enhanced robustness.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Deep Generative Models & Autoencoders
• Data Mining & Knowledge Management: Anomaly/Outlier Detection

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

Traditional deepfake detectors have dealt with the detection problem as a binary classification task. This approach can achieve satisfactory results in cases where samples of a given deepfake generation technique have been seen during training, but can easily fail with deepfakes generated by other techniques. In this paper, we propose DiffFake, a novel deepfake detector that approaches the detection problem as an anomaly detection task. Specifically, DiffFake learns natural changes that occur between two facial images of the same person by leveraging a differential anomaly detection framework. This is done by combining pairs of deep face embeddings and using them to train an anomaly detection model. We further propose to train a feature extractor on pseudo-deepfakes with global and local artifacts, to extract meaningful and generalizable features that can then be used to train the anomaly detection model. We perform extensive experiments on five different deepfake datasets and show that our method can match and sometimes even exceed the performance of state-of-the-art competitors.

Problem

• Detects deepfakes using differential anomaly detection
• Improves generalization across different deepfake techniques
• Trains model on pseudo-deepfakes with global and local artifacts

Innovation

• Differential anomaly detection framework
• Deep face embeddings combination
• Pseudo-deepfakes feature extraction