ML Security PapersDYMAPIA: A Multi-Domain Framework for Detecting AI-based Video Manipulation
Md Shohel Rana 1, Andrew H. Sung 2
Published on arXiv
2604.24426
Output Integrity Attack
OWASP ML Top 10 — ML09
Key Finding
Achieves over 99% accuracy and F1-score on FF++, Celeb-DF, and VDFD benchmarks while maintaining real-time inference capability
DYMAPIA
Novel technique introduced
AI-generated media are advancing rapidly, raising pressing concerns for content authenticity and digital trust. We introduce DYMAPIA, a multi-domain Deepfake detection framework that fuses spatial, spectral, and temporal cues to capture subtle traces of manipulation in visual data. The system builds dynamic anomaly masks by combining evidence from Fourier spectra, local texture descriptors, edge irregularities, and optical flow consistency, which highlight tampered regions with fine spatial accuracy. These masks guide DistXCNet, a lightweight classifier distilled from Xception and optimized with depthwise separable convolutions for fast, region-focused classification. This joint design achieves state-of-the-art results, with accuracy and F1-scores exceeding 99\% on FF++, Celeb-DF, and VDFD benchmarks, while keeping the model compact enough for real-time use. Beyond outperforming existing full-frame and multidomain detectors, DYMAPIA demonstrates deployment readiness for time-critical forensic tasks, including media verification, misinformation defense, and secure content filtering.
Key Contributions
- Multi-domain deepfake detection combining spatial, spectral, and temporal cues via dynamic anomaly masks
- Lightweight DistXCNet classifier distilled from Xception using depthwise separable convolutions for real-time deployment
- State-of-the-art performance exceeding 99% accuracy/F1 on FF++, Celeb-DF, and VDFD benchmarks
🛡️ Threat Analysis
Detects AI-generated and manipulated video content to verify output integrity and authenticity — core ML09 task of identifying synthetic media and deepfakes.