A Multi-Domain Feature Fusion Framework for Generalizable Deepfake Detection Across Different Generators

2026-06-12

AI Summary

This work addresses the limited generalization of existing deepfake detection methods when confronted with diverse generators such as diffusion models, as well as their inadequate exploitation of complementary multi-domain features and lack of robust cross-generator evaluation. To overcome these challenges, the authors propose SGFF-Net, which innovatively integrates spatial, gradient, and discrete wavelet transform (DWT) frequency-domain features within a dual residual learning architecture, enhanced by multi-source training and data augmentation strategies. The approach significantly improves detection generalization across both cross-model and cross-paradigm settings. Experimental results demonstrate state-of-the-art performance, achieving 98.95% accuracy in intra-dataset evaluations and 79.80%, 78.00%, and 75.80% in cross-model, cross-paradigm, and real-world scenarios, respectively, substantially outperforming current single-domain methods.

Abstract

Deepfakes are artificially generated images, audio, or videos that threaten privacy, security, and information integrity. Detecting such content is crucial for countering disinformation, as the latest models generate highly realistic content. While spatial- or frequency-based approaches achieve good detection rates on Generative Adversarial Networks (GANs)-based generated deepfakes, they often struggle with recent diffusion model-generated images. In particular, existing approaches rarely exploit complementary multi-domain representations or systematically evaluate cross-generator robustness. To address these challenges, we propose a multi-domain deepfake detection framework called SGFF-Net (Spatial-Gradient-Frequency Fusion Network) that integrates spatial, gradient, and DWT (Discrete Wavelet Transform)-based frequency representations within a dual residual learning architecture. Experimental results show that the SGFF-Net achieves 98.95\% accuracy in intra-dataset evaluation and improves performance in both cross-model (70.46\%) and cross-paradigm (69.94\%) settings. Incorporating multi-source training and data augmentation further enhances robustness, increasing accuracy from 70.46\% to 79.80\% in cross-model evaluation, from 69\% to 78\% in cross-paradigm evaluation, and from 61.50\% to 75.80\% on real-world data. Unlike single-domain detectors, the SGFF-Net learns complementary forensic cues across spatial, gradient, and wavelet-frequency domains, resulting in greater robustness under cross-generator and cross-paradigm evaluation. The results further show that combining multi-domain representations with data diversity and augmentation substantially improves generalization, providing practical insights for developing more reliable deepfake detection systems.

Problem

• deepfake detection
• cross-generator generalization
• multi-domain features
• diffusion models
• robustness

Innovation

• multi-domain fusion
• deepfake detection
• cross-generator generalization
• discrete wavelet transform
• dual residual learning