When Deepfake Detection Meets Graph Neural Network:a Unified and Lightweight Learning Framework

TL;DR

This paper presents SSTGNN, a lightweight graph neural network framework that effectively detects deepfakes by jointly analyzing spatial, spectral, and temporal cues, outperforming existing methods with fewer parameters.

Contribution

Introduces SSTGNN, a unified graph-based model that captures manipulation traces across multiple domains, significantly reducing model size while enhancing detection accuracy.

Findings

Achieves superior detection performance on benchmark datasets.

Uses up to 42 times fewer parameters than state-of-the-art models.

Demonstrates strong cross-domain generalization.

Abstract

The proliferation of generative video models has made detecting AI-generated and manipulated videos an urgent challenge. Existing detection approaches often fail to generalize across diverse manipulation types due to their reliance on isolated spatial, temporal, or spectral information, and typically require large models to perform well. This paper introduces SSTGNN, a lightweight Spatial-Spectral-Temporal Graph Neural Network framework that represents videos as structured graphs, enabling joint reasoning over spatial inconsistencies, temporal artifacts, and spectral distortions. SSTGNN incorporates learnable spectral filters and spatial-temporal differential modeling into a unified graph-based architecture, capturing subtle manipulation traces more effectively. Extensive experiments on diverse benchmark datasets demonstrate that SSTGNN not only achieves superior performance in both in-domain and cross-domain settings, but also offers strong efficiency and resource allocation. Remarkably, SSTGNN accomplishes these results with up to 42$\times$ fewer parameters than state-of-the-art models, making it highly lightweight and resource-friendly for real-world deployment.