Deepfake Forensics Adapter: A Dual-Stream Network for Generalizable Deepfake Detection

2026-03-01

AI Summary

This work addresses the limited generalization of current deepfake detection methods in the face of rapidly evolving forgery techniques. The authors propose a dual-stream network framework that, while keeping the parameters of a pretrained CLIP model frozen, effectively fuses its global semantic features with local anomaly cues—augmented by facial structural priors—through a Transformer-based interactive fusion mechanism to enhance detection of unseen forgery types. Evaluated on the DFDC dataset, the method achieves a frame-level AUC of 0.816 and a video-level AUC of 0.836, representing a 4.8% improvement in video-level AUC over the previous state-of-the-art, thereby significantly advancing cross-forgery-type generalization performance.

Technology Category

• Computer Vision: Adversarial Attacks & Robustness
• Machine Learning: Deep Generative Models & Autoencoders
• Natural Language Processing: Fact-Checking / Misinformation Detection (NLP Focus)

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

The rapid advancement of deepfake generation techniques poses significant threats to public safety and causes societal harm through the creation of highly realistic synthetic facial media. While existing detection methods demonstrate limitations in generalizing to emerging forgery patterns, this paper presents Deepfake Forensics Adapter (DFA), a novel dual-stream framework that synergizes vision-language foundation models with targeted forensics analysis. Our approach integrates a pre-trained CLIP model with three core components to achieve specialized deepfake detection by leveraging the powerful general capabilities of CLIP without changing CLIP parameters: 1) A Global Feature Adapter is used to identify global inconsistencies in image content that may indicate forgery, 2) A Local Anomaly Stream enhances the model's ability to perceive local facial forgery cues by explicitly leveraging facial structure priors, and 3) An Interactive Fusion Classifier promotes deep interaction and fusion between global and local features using a transformer encoder. Extensive evaluations of frame-level and video-level benchmarks demonstrate the superior generalization capabilities of DFA, particularly achieving state-of-the-art performance in the challenging DFDC dataset with frame-level AUC/EER of 0.816/0.256 and video-level AUC/EER of 0.836/0.251, representing a 4.8% video AUC improvement over previous methods. Our framework not only demonstrates state-of-the-art performance, but also points out a feasible and effective direction for developing a robust deepfake detection system with enhanced generalization capabilities against the evolving deepfake threats. Our code is available at https://github.com/Liao330/DFA.git

Problem

• deepfake detection
• generalization
• forgery patterns
• synthetic media
• forensics

Innovation

• Deepfake Detection
• Vision-Language Model
• Adapter Architecture
• Generalization
• Dual-Stream Network