A Comparison of Correlation Filter-Based Trackers and Struck Trackers

doi:10.1109/TCSVT.2019.2931924

CORC > 自动化研究所 > 中国科学院自动化研究所 > 模式识别国家重点实验室 > 图像与视频分析团队

	A Comparison of Correlation Filter-Based Trackers and Struck Trackers
	Wang, Jinqiao 1,2; Zheng, Linyu 1,2; Tang, Ming 1,2; Feng, Jiayi 3
刊名	IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
	2020-09-01
卷号	30 期号:9 页码:3106-3118
关键词	Correlation Target tracking Kernel Support vector machines Training Optimization Discrete Fourier transforms Visual tracking correlation filters structured output SVM tracker struck ranking SVM tracker
ISSN号	1051-8215
DOI	10.1109/TCSVT.2019.2931924
通讯作者	Wang, Jinqiao(jqwang@nlpr.ia.ac.cn)
英文摘要	In recent years, two types of trackers, namely correlation filter-based tracker (CF tracker) and structured output tracker (struck), have exhibited the state-of-the-art performance. However, there seems to be a lack of analytic work on their relations in the computer vision community. In this paper, we investigate two state-of-the-art CF trackers, i.e., spatial regularization discriminative correlation filter (SRDCF) and correlation filter with limited boundaries (CFLB), and struck, and reveal their relations. Specifically, after extending the CFLB to its multiple channel versions, we prove the relation between SRDCF and CFLB on the condition that the spatial regularization factor of SRDCF is replaced by the masking matrix of CFLB. We also prove the asymptotical approximate relation between SRDCF and struck on the conditions that the spatial regularization factor of SRDCF is replaced by an indicator function of object bounding box, the weights of SRDCF in its loss item are replaced by those of struck, the linear kernel is employed by struck, and the search region tends to infinity. The extensive experiments on public benchmarks OTB50 and OTB100 are conducted to verify our theoretical results. Moreover, we explain how detailed differences among SRDCF, CFLB, and Struck would give rise to slightly different performances on visual sequences.
资助项目	Natural Science Foundation of China[61772527]
WOS关键词	OBJECT TRACKING ; VISUAL TRACKING
WOS研究方向	Engineering
语种	英语
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
WOS记录号	WOS:000567499300025
资助机构	Natural Science Foundation of China
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/41953]
专题	自动化研究所_模式识别国家重点实验室_图像与视频分析团队
通讯作者	Wang, Jinqiao
作者单位	1.Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, Peoples R China 2.Univ Chinese Acad Sci, Beijing 100049, Peoples R China 3.Microsoft Search Technol Ctr Asia STCA, Beijing, Peoples R China
推荐引用方式 GB/T 7714	Wang, Jinqiao,Zheng, Linyu,Tang, Ming,et al. A Comparison of Correlation Filter-Based Trackers and Struck Trackers[J]. IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY,2020,30(9):3106-3118.
APA	Wang, Jinqiao,Zheng, Linyu,Tang, Ming,&Feng, Jiayi.(2020).A Comparison of Correlation Filter-Based Trackers and Struck Trackers.IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY,30(9),3106-3118.
MLA	Wang, Jinqiao,et al."A Comparison of Correlation Filter-Based Trackers and Struck Trackers".IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY 30.9(2020):3106-3118.