DYMAPIA: A Multi-Domain Framework for Detecting AI-based Video Manipulation

TL;DR

DYMAPIA is a multi-domain deepfake detection framework that combines spatial, spectral, and temporal cues to accurately identify manipulated videos in real-time.

Contribution

It introduces a novel multi-domain approach with dynamic anomaly masks and a lightweight classifier, achieving state-of-the-art accuracy while enabling real-time forensic applications.

Findings

Achieves over 99% accuracy and F1-score on multiple benchmarks.

Outperforms existing full-frame and multidomain detectors.

Enables deployment in time-critical forensic scenarios.

Abstract

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.