Zero-Shot Fake Video Detection by Audio-Visual Consistency

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Li, Xiaolou, Liu, Zehua, Chen, Chen, Li, Lantian, Guo, Li, Wang, Dong
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914832473653248
author Li, Xiaolou
Liu, Zehua
Chen, Chen
Li, Lantian
Guo, Li
Wang, Dong
author_facet Li, Xiaolou
Liu, Zehua
Chen, Chen
Li, Lantian
Guo, Li
Wang, Dong
contents Recent studies have advocated the detection of fake videos as a one-class detection task, predicated on the hypothesis that the consistency between audio and visual modalities of genuine data is more significant than that of fake data. This methodology, which solely relies on genuine audio-visual data while negating the need for forged counterparts, is thus delineated as a `zero-shot' detection paradigm. This paper introduces a novel zero-shot detection approach anchored in content consistency across audio and video. By employing pre-trained ASR and VSR models, we recognize the audio and video content sequences, respectively. Then, the edit distance between the two sequences is computed to assess whether the claimed video is genuine. Experimental results indicate that, compared to two mainstream approaches based on semantic consistency and temporal consistency, our approach achieves superior generalizability across various deepfake techniques and demonstrates strong robustness against audio-visual perturbations. Finally, state-of-the-art performance gains can be achieved by simply integrating the decision scores of these three systems.
format Preprint
id arxiv_https___arxiv_org_abs_2406_07854
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Zero-Shot Fake Video Detection by Audio-Visual Consistency
Li, Xiaolou
Liu, Zehua
Chen, Chen
Li, Lantian
Guo, Li
Wang, Dong
Sound
Multimedia
Audio and Speech Processing
Recent studies have advocated the detection of fake videos as a one-class detection task, predicated on the hypothesis that the consistency between audio and visual modalities of genuine data is more significant than that of fake data. This methodology, which solely relies on genuine audio-visual data while negating the need for forged counterparts, is thus delineated as a `zero-shot' detection paradigm. This paper introduces a novel zero-shot detection approach anchored in content consistency across audio and video. By employing pre-trained ASR and VSR models, we recognize the audio and video content sequences, respectively. Then, the edit distance between the two sequences is computed to assess whether the claimed video is genuine. Experimental results indicate that, compared to two mainstream approaches based on semantic consistency and temporal consistency, our approach achieves superior generalizability across various deepfake techniques and demonstrates strong robustness against audio-visual perturbations. Finally, state-of-the-art performance gains can be achieved by simply integrating the decision scores of these three systems.
title Zero-Shot Fake Video Detection by Audio-Visual Consistency
topic Sound
Multimedia
Audio and Speech Processing
url https://arxiv.org/abs/2406.07854