Research Article
Face Forgery Detection with Long-Range Noise Features and Multilevel Frequency-Aware Clues
Figure 3
The pipeline of the proposed framework. We design a dual-branch collaborative learning framework to comprehensively dig into the long-range noise features and multilevel frequency-aware clues. ANTEM represents the adaptive noise trace enhancement module. MFAM represents the multilevel frequency-aware module. The collaboration strategy of single center loss and cross-entropy loss supervise the framework to learn more generalized and robust features.