Transformations as Denoising: A Robust Approach to Weaken Adversarial Facial Images
Author:
Affiliation:
1. Towson University,Dept. of Computer and Information Science,USA
2. MOVEJ Analytics,USA
Publisher
IEEE
Link
http://xplorestaging.ieee.org/ielx7/9925246/9925275/09925338.pdf?arnumber=9925338
Reference47 articles.
1. One Pixel Attack for Fooling Deep Neural Networks
2. The Limitations of Deep Learning in Adversarial Settings
3. The Unreasonable Effectiveness of Deep Features as a Perceptual Metric
4. Towards transferable adversarial attack against deep face recognition;zhong;IEEE Transactions on Information Forensics and Security,2020
5. Backdoor Attacks Against Deep Learning Systems in the Physical World
Cited by 2 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. Deblurring as a Defense against Adversarial Attacks;2023 IEEE 12th International Conference on Cloud Networking (CloudNet);2023-11-01
2. Performance of GAN-Based Denoising and Restoration Techniques for Adversarial Face Images;2023 IEEE/ACIS 21st International Conference on Software Engineering Research, Management and Applications (SERA);2023-05-23
1.学者识别学者识别
2.学术分析学术分析
3.人才评估人才评估
"同舟云学术"是以全球学者为主线，采集、加工和组织学术论文而形成的新型学术文献查询和分析系统，可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容，当前同舟云学术共收录了国内外主流学术期刊6万余种，收集的期刊论文及会议论文总量共计约1.5亿篇，并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询！咨询电话：010-8811{复制后删除}0370
www.globalauthorid.com
TOP
Copyright © 2019-2024 北京同舟云网络信息技术有限公司 京公网安备11010802033243号 京ICP备18003416号-3