Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

CAM-VFD: Cross-Attention Multimodal Video Forgery Detection

About

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}.

Hoda Osama Elkhodary, Sherin Mostafa Youssef, Marwa Elshenawy, Dalia Sobhy• 2026

Related benchmarks

TaskDatasetResultRank
AI-generated Video DetectionGenVideo Show1
ACC95.1
12
AI-generated Video DetectionGenVideo Wild Scrape
Accuracy92.3
12
Video Forgery DetectionGenVidBench
HD-VG Score99.52
12
AI-generated Video DetectionGenVideo Morph Studio
Accuracy92.5
12
AI-generated Video DetectionGenVideo Sora
Accuracy89.5
12
AI-generated Video DetectionGenVideo ModelScope
Accuracy94.8
12
AI-generated Video DetectionGenVideo Lavie
ACC93.6
12
AI-generated Video DetectionGenVideo Gen2
Accuracy93.2
12
AI-generated Video DetectionGenVideo Moon Valley
Accuracy90.6
12
Generated Video DetectionGenVideo Hot Shot
Recall91.8
6
Showing 10 of 11 rows

Other info

Follow for update