Online selection of the best k-feature subset for object tracking

doi:10.1016/j.jvcir.2011.11.001

	Online selection of the best k-feature subset for object tracking
	Li, Guorong 1; Huang, Qingming 1,2; Pang, Junbiao 2; Jiang, Shuqiang 2; Qin, Lei 2
	2012-02-01
发表期刊	JOURNAL OF VISUAL COMMUNICATION AND IMAGE REPRESENTATION
ISSN	1047-3203
卷号	23 期号:2 页码:254-263
摘要	In this paper, we propose a new feature subset evaluation method for feature selection in object tracking. According to the fact that a feature which is useless by itself could become a good one when it is used together with some other features, we propose to evaluate feature subsets as a whole for object tracking instead of scoring each feature individually and find out the most distinguishable subset for tracking. In the paper, we use a special tree to formalize the feature subset space. Then conditional entropy is used to evaluating feature subset and a simple but efficient greedy search algorithm is developed to search this tree to obtain the optimal k-feature subset quickly. Furthermore, our online k-feature subset selection method is integrated into particle filter for robust tracking. Extensive experiments demonstrate that k-feature subset selected by our method is more discriminative and thus can improve tracking performance considerably. (C) 2011 Elsevier Inc. All rights reserved.
关键词	Object tracking Feature subset selection Feature selection Feature subset tree Conditional entropy Greedy search algorithm Particle filter Online selection
DOI	10.1016/j.jvcir.2011.11.001
收录类别	SCI
语种	英语
资助项目	National Basic Research Program of China (973 Program)[2009CB320906] ; National Natural Science Foundation of China[61025011] ; National Natural Science Foundation of China[61035001] ; National Natural Science Foundation of China[61133003] ; National Natural Science Foundation of China[61003165] ; Beijing Natural Science Foundation[4111003]
WOS研究方向	Computer Science
WOS类目	Computer Science, Information Systems ; Computer Science, Software Engineering
WOS记录号	WOS:000300198900003
出版者	ACADEMIC PRESS INC ELSEVIER SCIENCE
引用统计	被引频次：8[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/2XEOYT63/5380
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Li, Guorong
作者单位	1.Chinese Acad Sci, Grad Univ, Beijing 100190, Peoples R China 2.CAS, Inst Comput Tech, Key Lab Intell Info Proc, Beijing 100080, Peoples R China
推荐引用方式 GB/T 7714	Li, Guorong,Huang, Qingming,Pang, Junbiao,et al. Online selection of the best k-feature subset for object tracking[J]. JOURNAL OF VISUAL COMMUNICATION AND IMAGE REPRESENTATION,2012,23(2):254-263.
APA	Li, Guorong,Huang, Qingming,Pang, Junbiao,Jiang, Shuqiang,&Qin, Lei.(2012).Online selection of the best k-feature subset for object tracking.JOURNAL OF VISUAL COMMUNICATION AND IMAGE REPRESENTATION,23(2),254-263.
MLA	Li, Guorong,et al."Online selection of the best k-feature subset for object tracking".JOURNAL OF VISUAL COMMUNICATION AND IMAGE REPRESENTATION 23.2(2012):254-263.