Author:
Hao Hanxiang,Bartusiak Emily R.,Güera David,Mas Montserrat Daniel,Baireddy Sriram,Xiang Ziyue,Yarlagadda Sri Kalyan,Shao Ruiting,Horváth János,Yang Justin,Zhu Fengqing,Delp Edward J.
Abstract
AbstractFalsified media threatens key areas of our society, ranging from politics to journalism to economics. Simple and inexpensive tools available today enable easy, credible manipulations of multimedia assets. Some even utilize advanced artificial intelligence concepts to manipulate media, resulting in videos known asdeepfakes. Social media platforms and their “echo chamber” effect propagate fabricated digital content at scale, sometimes with dire consequences in real-world situations. However, ensuring semantic consistency across falsified media assets of different modalities is still very challenging for current deepfake tools. Therefore, cross-modal analysis (e.g., video-based and audio-based analysis) provides forensic analysts an opportunity to identify inconsistencies with higher accuracy. In this chapter, we introduce several approaches to detect deepfakes. These approaches leverage different data modalities, including video and audio. We show that the presented methods achieve accurate detection for various large-scale datasets.
Publisher
Springer International Publishing
Cited by
5 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
1. Attention-based Multimodal learning framework for Generalized Audio- Visual Deepfake Detection;2023-10-11
2. Transformer-Based Speech Synthesizer Attribution in an Open Set Scenario;2022 21st IEEE International Conference on Machine Learning and Applications (ICMLA);2022-12
3. Comparative Analysis on Different DeepFake Detection Methods and Semi Supervised GAN Architecture for DeepFake Detection;2022 Sixth International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud) (I-SMAC);2022-11-10
4. An Overview of Recent Work in Multimedia Forensics;2022 IEEE 5th International Conference on Multimedia Information Processing and Retrieval (MIPR);2022-08
5. Multi-model DeepFake Detection Using Deep and Temporal Features;Third International Conference on Image Processing and Capsule Networks;2022