AdaForensics: Learning A Characteristic-aware Adaptive Deepfake Detector

2026-08-03

AI Summary

This work addresses the limitations of existing deepfake detection methods, which rely on fixed models and struggle to generalize across diverse facial characteristics. To overcome this, the authors propose AdaForensics, a novel framework that introduces, for the first time, a feature-aware dynamic adaptation mechanism. Leveraging a dual-branch hypernetwork architecture, AdaForensics simultaneously learns both face-agnostic and face-specific embeddings, enabling it to dynamically adjust detector parameters at test time. This design effectively balances generalization and personalization in forgery detection. Extensive experiments demonstrate that AdaForensics significantly outperforms state-of-the-art methods across multiple benchmark datasets, including FaceForensics, Celeb-DF, and DFDC.

Abstract

In this paper, we propose a characteristic-aware adaptive network named AdaForensics for deepfake detection. Most existing methods learn a fixed network to detect deepfakes based on carefully-designed network architectures. However, these methods employ the same deepfake detector for all the images despite of various facial characteristic, which fail to provide customized forgery detection for different individuals. To address this, our AdaForensics simultaneously learns characteristic-agnostic and characteristic-specific embeddings, where the detector dynamically adapts to varying faces with our designed hypernetwork on the fly. More specifically, our AdaForensics not only explores the shareable abstractions from various deepfake images, but also adapts the detector to the given characteristic at test time. To achieve this, we propose a two-branch HyperNetwork to learn an adaptive deepfake detector, which automatically adjusts the parameters based on characteristic of the input. Extensive experiments on widely-used datasets including FaceForensics, Celeb-DF and DFDC demonstrate our AdaForensics outperforms the state-of-the-art works.

Problem

• deepfake detection
• facial characteristics
• adaptive detection
• forgery detection
• individual-specific

Innovation

• adaptive deepfake detection
• characteristic-aware
• HyperNetwork
• dynamic parameter adaptation
• face forensics