Face-D(^2)CL: Multi-Domain Synergistic Representation with Dual Continual Learning for Facial DeepFake Detection

TL;DR

This paper introduces Face-D(^2)CL, a novel multi-domain, dual continual learning framework that enhances facial DeepFake detection by combining spatial and frequency features and addressing catastrophic forgetting.

Contribution

It proposes a synergistic multi-domain feature fusion and a dual continual learning mechanism with EWC and OGC to improve detection robustness and adaptability without historical data replay.

Findings

Achieves 60.7% relative reduction in detection error rate.

Outperforms SOTA in stability and plasticity.

Improves detection AUC by 7.9% on unseen forgery domains.

Abstract

The rapid advancement of facial forgery techniques poses severe threats to public trust and information security, making facial DeepFake detection a critical research priority. Continual learning provides an effective approach to adapt facial DeepFake detection models to evolving forgery patterns. However, existing methods face two key bottlenecks in real-world continual learning scenarios: insufficient feature representation and catastrophic forgetting. To address these issues, we propose Face-D(^2)CL, a framework for facial DeepFake detection. It leverages multi-domain synergistic representation to fuse spatial and frequency-domain features for the comprehensive capture of diverse forgery traces, and employs a dual continual learning mechanism that combines Elastic Weight Consolidation (EWC), which distinguishes parameter importance for real versus fake samples, and Orthogonal Gradient Constraint (OGC), which ensures updates to task-specific adapters do not interfere with previously learned knowledge. This synergy enables the model to achieve a dynamic balance between robust anti-forgetting capabilities and agile adaptability to emerging facial forgery paradigms, all without relying on historical data replay. Extensive experiments demonstrate that our method surpasses current SOTA approaches in both stability and plasticity, achieving 60.7% relative reduction in average detection error rate, respectively. On unseen forgery domains, it further improves the average detection AUC by 7.9% compared to the current SOTA method.