Affiliation:
1. Dipartimento di Elettronica, Informazione e Bioingegneria, Politecnico di Milano, 20133 Milan, Italy
2. School of Cyber Science and Technology, University of Science and Technology of China, Hefei 230026, China
Abstract
The widespread use of deep learning techniques for creating realistic synthetic media, commonly known as deepfakes, poses a significant threat to individuals, organizations, and society. As the malicious use of these data could lead to unpleasant situations, it is becoming crucial to distinguish between authentic and fake media. Nonetheless, though deepfake generation systems can create convincing images and audio, they may struggle to maintain consistency across different data modalities, such as producing a realistic video sequence where both visual frames and speech are fake and consistent one with the other. Moreover, these systems may not accurately reproduce semantic and timely accurate aspects. All these elements can be exploited to perform a robust detection of fake content. In this paper, we propose a novel approach for detecting deepfake video sequences by leveraging data multimodality. Our method extracts audio-visual features from the input video over time and analyzes them using time-aware neural networks. We exploit both the video and audio modalities to leverage the inconsistencies between and within them, enhancing the final detection performance. The peculiarity of the proposed method is that we never train on multimodal deepfake data, but on disjoint monomodal datasets which contain visual-only or audio-only deepfakes. This frees us from leveraging multimodal datasets during training, which is desirable given their lack in the literature. Moreover, at test time, it allows to evaluate the robustness of our proposed detector on unseen multimodal deepfakes. We test different fusion techniques between data modalities and investigate which one leads to more robust predictions by the developed detectors. Our results indicate that a multimodal approach is more effective than a monomodal one, even if trained on disjoint monomodal datasets.
Funder
Defense Advanced Research Projects Agency
European Union
Subject
Electrical and Electronic Engineering,Computer Graphics and Computer-Aided Design,Computer Vision and Pattern Recognition,Radiology, Nuclear Medicine and imaging
Reference57 articles.
1. The New York Times (The New York Times, 2022). Science Has a Nasty Photoshopping Problem, The New York Times.
2. VICE (VICE, 2023). How I Broke Into a Bank Account With an AI-Generated Voice, VICE.
3. The Verge (The Verge, 2022). Liveness Tests Used by Banks to Verify ID Are “Extremely Vulnerable” to Deepfake Attacks, The Verge.
4. Media forensics and deepfakes: An overview;Verdoliva;IEEE J. Sel. Top. Signal Process.,2020
5. Deepfakes Generation and Detection: State-of-the-art, open challenges, countermeasures, and way forward;Masood;Appl. Intell.,2022
Cited by
14 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献