2025/10/31 by Hyemin Boo, Boo, Hyemin, Eunsang Lee +3
Computer Science · #Code (set theory) #Computer Vision and Pattern Recognition (cs.CV) #Digital Media Forensic Detection #FOS: Computer and information sciences #Generative Adversarial Networks and Image Synthesis #Generative grammar #Generative model #Identity (music) #Matching (statistics) #Multimedia (cs.MM) #Speech and Audio Processing
paper · pdf · doi:10.48550/arxiv.2510.27475
published in arXiv (Cornell University) (Cornell University)
openalex publication_date 2025/10/31 · openalex created_date 2025/11/05 · openalex updated_date 2026/08/05
Deepfakes generated by advanced generative models have rapidly posed serious threats, yet existing audiovisual deepfake detection approaches struggle to generalize to unseen manipulation methods. To address this, we propose a novel reference-aware audiovisual deepfake detection method, called Referee to capture fine-grained identity discrepancies. Unlike existing methods that overfit to transient spatiotemporal artifacts, Referee employs identity bottleneck and matching modules to model the relational consistency of speaker-specific cues captured by a single one-shot example as a biometric anchor. Extensive experiments on FakeAVCeleb, FaceForensics++, and KoDF demonstrate that Referee achieves state-of-the-art results on cross-dataset and cross-language evaluation protocols, including a 99.4% AUC on KoDF. These results highlight that explicitly correlating reference-based biometric priors is a key frontier for achieving generalized and reliable audiovisual forensics. The code is available at https://github.com/ewha-mmai/referee.