{"title":"Adversarial Attacks Against LipNet: End-to-End Sentence Level Lipreading","authors":"Mahir Jethanandani, Derek Tang","doi":"10.1109/SPW50608.2020.00020","DOIUrl":null,"url":null,"abstract":"Visual adversarial attacks inspired by Carlini-Wagner targeted audiovisual attacks can fool the state-of-the-art Google DeepMind LipNet model to subtitle anything with over 99% similarity. We explore several methods of visual adversarial attacks, including the vanilla fast gradient sign method (FGSM), the $L_{\\infty}$ iterative fast gradient sign method, and the $L_{2}$ modified Carlini-Wagner attacks. The feasibility of these attacks raise privacy and false information threats, as video transcriptions are used to recommend and inform people worldwide and on social media.","PeriodicalId":413600,"journal":{"name":"2020 IEEE Security and Privacy Workshops (SPW)","volume":"52 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2020-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"7","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2020 IEEE Security and Privacy Workshops (SPW)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/SPW50608.2020.00020","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 7
Abstract
Visual adversarial attacks inspired by Carlini-Wagner targeted audiovisual attacks can fool the state-of-the-art Google DeepMind LipNet model to subtitle anything with over 99% similarity. We explore several methods of visual adversarial attacks, including the vanilla fast gradient sign method (FGSM), the $L_{\infty}$ iterative fast gradient sign method, and the $L_{2}$ modified Carlini-Wagner attacks. The feasibility of these attacks raise privacy and false information threats, as video transcriptions are used to recommend and inform people worldwide and on social media.
受Carlini-Wagner启发的视觉对抗攻击可以欺骗最先进的谷歌DeepMind LipNet模型,使其为任何超过99的内容添加字幕% similarity. We explore several methods of visual adversarial attacks, including the vanilla fast gradient sign method (FGSM), the $L_{\infty}$ iterative fast gradient sign method, and the $L_{2}$ modified Carlini-Wagner attacks. The feasibility of these attacks raise privacy and false information threats, as video transcriptions are used to recommend and inform people worldwide and on social media.