FakeRadar: Probing Forgery Outliers to Detect Unknown Deepfake Videos

2025-12-16

AI Summary

To address the weak generalization and over-reliance on known manipulation patterns in detecting unseen deepfake videos, this paper proposes a prior-agnostic universal detection framework. Methodologically, it introduces: (1) a novel forged anomaly probing mechanism that actively identifies distributional gaps in feature space to detect boundary anomalies among authentic, known-forged, and unseen-forged samples; (2) dynamic sub-clustering modeling coupled with boundary-proximal anomalous sample generation; and (3) an anomaly-guided three-stage training paradigm integrating anomaly-driven contrastive learning and cluster-conditioned cross-entropy loss. Leveraging foundation models such as CLIP, the method achieves state-of-the-art performance across multiple benchmarks and cross-domain evaluations. It demonstrates strong generalization to emerging forgery techniques, significantly advancing detection robustness and setting new performance benchmarks.

Technology Category

• Computer Vision: Adversarial Attacks & Robustness
• Machine Learning: Deep Neural Architectures and Foundation Models
• Data Mining & Knowledge Management: Anomaly/Outlier Detection

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

In this paper, we propose FakeRadar, a novel deepfake video detection framework designed to address the challenges of cross-domain generalization in real-world scenarios. Existing detection methods typically rely on manipulation-specific cues, performing well on known forgery types but exhibiting severe limitations against emerging manipulation techniques. This poor generalization stems from their inability to adapt effectively to unseen forgery patterns. To overcome this, we leverage large-scale pretrained models (e.g. CLIP) to proactively probe the feature space, explicitly highlighting distributional gaps between real videos, known forgeries, and unseen manipulations. Specifically, FakeRadar introduces Forgery Outlier Probing, which employs dynamic subcluster modeling and cluster-conditional outlier generation to synthesize outlier samples near boundaries of estimated subclusters, simulating novel forgery artifacts beyond known manipulation types. Additionally, we design Outlier-Guided Tri-Training, which optimizes the detector to distinguish real, fake, and outlier samples using proposed outlier-driven contrastive learning and outlier-conditioned cross-entropy losses. Experiments show that FakeRadar outperforms existing methods across various benchmark datasets for deepfake video detection, particularly in cross-domain evaluations, by handling the variety of emerging manipulation techniques.

Problem

• Detect unknown deepfake videos in cross-domain scenarios
• Overcome poor generalization to unseen forgery patterns
• Probe feature space to highlight distributional gaps

Innovation

• Probes feature space using pretrained models
• Generates outlier samples via dynamic subcluster modeling
• Employs outlier-guided tri-training with contrastive learning