Contour-motion feature (CMF): A space-time approach for robust pedestrian detection

doi:10.1016/j.patrec.2008.03.007

	Contour-motion feature (CMF): A space-time approach for robust pedestrian detection
	Liu, Yazhou 1; Chen, Xilin 2; Yao, Hongxun 1; Cui, Xinyi 1; Liu, Chaoran 2; Gao, Wen 2
	2009-01-15
发表期刊	PATTERN RECOGNITION LETTERS
ISSN	0167-8655
卷号	30 期号:2 页码:148-156
摘要	This paper presents a contour-motion feature for robust pedestrian detection. The space-time contours are used as the low level representation of the pedestrian. Then we apply 3D distance transform to extend the I-dimensional contour into 3-dimensional space. By this way, the relations between the local contours can be maintained implicitly. Further, by encapsulating the static and dynamic information by 3D Haar-like filters, we can generate the middle level pedestrian representation: contour-motion features. Then we use boosting method to select the most representative features. Our experiments demonstrate that the proposed approach can Outperform Viola's well-known pedestrian detector in both detection accuracy and generalization ability. In addition, even though our approach is presented in pedestrian detection scenario, it has been extended to human activity recognition application and remarkable performance has been achieved. (C) 2008 Elsevier B.V. All rights reserved.
关键词	Contour-motion feature Space-time analysis Pedestrian detection Human activity analysis Event analysis 3D Haar filter
DOI	10.1016/j.patrec.2008.03.007
收录类别	SCI
语种	英语
资助项目	National Natural Science Foundation of China[60332010] ; National Natural Science Foundation of China[60673091] ; National Natural Science Foundation of China[60772071] ; Hi-Tech Research and Development Program of China[2006AA01Z122] ; Hi-Tech Research and Development Program of China[2007AA01Z163] ; Program of New Century Excellent Talents in University[NCET-04-0320] ; Natural Science Foundation of Beijing municipal[4061001] ; Natural Science Foundation of Beijing municipal[4072023] ; ISVISION Technology Co. Ltd
WOS研究方向	Computer Science
WOS类目	Computer Science, Artificial Intelligence
WOS记录号	WOS:000261857200009
出版者	ELSEVIER SCIENCE BV
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/2XEOYT63/11680
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Liu, Yazhou
作者单位	1.Harbin Inst Technol, Sch Comp Sci & Technol, Harbin 150001, Peoples R China 2.Chinese Acad Sci, Inst Comp Technol, Beijing 100080, Peoples R China
推荐引用方式 GB/T 7714	Liu, Yazhou,Chen, Xilin,Yao, Hongxun,et al. Contour-motion feature (CMF): A space-time approach for robust pedestrian detection[J]. PATTERN RECOGNITION LETTERS,2009,30(2):148-156.
APA	Liu, Yazhou,Chen, Xilin,Yao, Hongxun,Cui, Xinyi,Liu, Chaoran,&Gao, Wen.(2009).Contour-motion feature (CMF): A space-time approach for robust pedestrian detection.PATTERN RECOGNITION LETTERS,30(2),148-156.
MLA	Liu, Yazhou,et al."Contour-motion feature (CMF): A space-time approach for robust pedestrian detection".PATTERN RECOGNITION LETTERS 30.2(2009):148-156.