CAM-VFD: Cross-Attention Multimodal Video Forgery Detection

TL;DR

CAM-VFD introduces a cross-attention multimodal framework leveraging cross-modal contradictions for robust video forgery detection, outperforming single-modality detectors and maintaining stability under various perturbations.

Contribution

It proposes a novel cross-attention fusion mechanism that models cross-modal contradictions for improved video forgery detection, demonstrating superior accuracy and robustness.

Findings

95.31% Top-1 accuracy on GenVidBench

93.43% accuracy and 90.63% F1-score on GenVideo

Stable performance under compression, noise, blur, and adversarial attacks

Abstract

The rapid advancement of Deepfake technologies and video manipulation tools poses a critical challenge to multimedia forensics, judicial evidence integrity, and information authenticity. Current detectors rely on single-modality signals, treating appearance, geometry, and motion independently. However, advanced generators maintain within-modality consistency while producing cross-modal contradictions, which are forensically discriminative but invisible to any single-modal detector. We propose CAM-VFD, a Cross-Attention Multimodal Video Forgery Detection framework that models cross-modal contradiction as a directional forensic signal. The framework uses a cross-attention fusion mechanism in which CLIP-based appearance representations serve as queries against VideoMAE motion features and MiDaS depth features, enabling the identification of contradictions between visual, temporal, and geometric evidence. We examine this design through cross-modal attention discrepancy analysis, observing statistically separable real and fake distributions ($p<0.001$, Cohen's $d=0.68$). Experimental results on two generative video benchmarks indicate consistent performance, with 95.31\% Top-1 accuracy on GenVidBench and 93.43\% accuracy, 90.63\% F1-score, and 96.56\% AUROC on GenVideo. Moreover, CAM-VFD demonstrates stable performance under compression, noise, blur, and adversarial perturbations, suggesting that cross-modal reasoning may improve robustness in media forensics. The code is publicly available at \url{https://github.com/Hoda-Osama/CAM-VFD/tree/main}.