Real-Time Robust Video Object Detection System Against Physical-World Adversarial Attacks

doi:10.1109/TCAD.2023.3305932

	Real-Time Robust Video Object Detection System Against Physical-World Adversarial Attacks
	Han, Husheng 1,2,3; Hu, Xing 1,4; Hao, Yifan 3; Xu, Kaidi 5; Dang, Pucheng 1,2,3; Wang, Ying 6; Zhao, Yongwei 7; Du, Zidong 1,4; Guo, Qi ; Wang, Yanzhi 6; Zhang, Xishan 1,7; Chen, Tianshi 8
	2024
发表期刊	IEEE TRANSACTIONS ON COMPUTER-AIDED DESIGN OF INTEGRATED CIRCUITS AND SYSTEMS
ISSN	0278-0070
卷号	43 期号:1 页码:366-379
摘要	DNN-based video object detection (VOD) powers autonomous driving and video surveillance industries with rising importance and promising opportunities. However, adversarial patch attack yields huge concern in live vision tasks because of its practicality, feasibility, and powerful attack effectiveness. This work proposes Themis, a software/hardware system to defend against adversarial patches for real-time robust VOD. We observe that adversarial patches exhibit extremely localized superficial feature importance in a small region with nonrobust predictions, and thus propose the adversarial region detection algorithm for adversarial effect elimination. Themis also proposes a systematic design to efficiently support the algorithm by eliminating redundant computations and memory traffics. Experimental results show that the proposed methodology can effectively recover the system from the adversarial attack with negligible hardware overhead.
关键词	Object detection Streaming media Optical flow Feature extraction Real-time systems Task analysis Detectors Adversarial patch attack deep learning security domain-specific accelerator hardware/software co-design real time
DOI	10.1109/TCAD.2023.3305932
收录类别	SCI
语种	英语
资助项目	National Key Research and Development Program of China
WOS研究方向	Computer Science ; Engineering
WOS类目	Computer Science, Hardware & Architecture ; Computer Science, Interdisciplinary Applications ; Engineering, Electrical & Electronic
WOS记录号	WOS:001129816700018
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
引用统计	被引频次：1[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/38428
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Hu, Xing
作者单位	1.Chinese Acad Sci, Inst Comp Technol, State Key Lab Processors, Beijing 100190, Peoples R China 2.Univ Chinese Acad Sci, Sch Comp Sci, Beijing 100049, Peoples R China 3.Cambricon Technol, Dept Architecture Algorithm, Beijing 100191, Peoples R China 4.Chinese Acad Sci, Shanghai Innovat Ctr Processor Technol, Beijing 100190, Peoples R China 5.Drexel Univ, Coll Comp & Informat, Dept Comp Sci, Philadelphia, PA 19104 USA 6.Northeastern Univ, Dept Elect & Comp Engn, Boston, MA 02115 USA 7.Cambricon Technol, Dept Architecture Algorithm, Beijing 100191, Peoples R China 8.Cambricon Technol, Beijing 100191, Peoples R China
推荐引用方式 GB/T 7714	Han, Husheng,Hu, Xing,Hao, Yifan,et al. Real-Time Robust Video Object Detection System Against Physical-World Adversarial Attacks[J]. IEEE TRANSACTIONS ON COMPUTER-AIDED DESIGN OF INTEGRATED CIRCUITS AND SYSTEMS,2024,43(1):366-379.
APA	Han, Husheng.,Hu, Xing.,Hao, Yifan.,Xu, Kaidi.,Dang, Pucheng.,...&Chen, Tianshi.(2024).Real-Time Robust Video Object Detection System Against Physical-World Adversarial Attacks.IEEE TRANSACTIONS ON COMPUTER-AIDED DESIGN OF INTEGRATED CIRCUITS AND SYSTEMS,43(1),366-379.
MLA	Han, Husheng,et al."Real-Time Robust Video Object Detection System Against Physical-World Adversarial Attacks".IEEE TRANSACTIONS ON COMPUTER-AIDED DESIGN OF INTEGRATED CIRCUITS AND SYSTEMS 43.1(2024):366-379.