SFANet: Spatial-Frequency Attention Network for Deepfake Detection

Detecting manipulated media has now become a pressing issue with the recent rise of deepfakes. Most existing approaches fail to generalize across diverse datasets and generation techniques. We thus propose a novel ensemble framework, combining the strengths of transformer-based architectures, such as Swin Transformers and ViTs, and texture-based methods, to achieve better detection accuracy and robustness. Our method introduces innovative data-splitting, sequential training, frequency splitting, patch-based attention, and face segmentation techniques to handle dataset imbalances, enhance high-impact regions (e.g., eyes and mouth), and improve generalization. Our model achieves state-of-the-art performance when tested on the DFWild-Cup dataset, a diverse subset of eight deepfake datasets. The ensemble benefits from the complementarity of these approaches, with transformers excelling in global feature extraction and texturebased methods providing interpretability. This work demonstrates that hybrid models can effectively address the evolving challenges of deepfake detection, offering a robust solution for real-world applications.

Key Contributions

• Ensemble framework combining Swin Transformers, ViTs, and texture-based (LBP/FFT) methods for complementary global and local feature extraction
• Novel pipeline techniques including frequency splitting, patch-based attention on high-impact facial regions (eyes, mouth), and face segmentation to improve generalization
• Sequential training and data-splitting strategies to handle dataset imbalance across diverse deepfake generation methods

🛡️ Threat Analysis

Details

Similar Papers