Beyond appearance model: Learning appearance variations for object tracking

doi:10.1016/j.neucom.2016.06.058

	Beyond appearance model: Learning appearance variations for object tracking
	Li, Guorong 1,2,3; Ma, Bingpeng 1,2,3; Huang, Jun 1; Huang, Qingming 1,2,3; Zhang, Weigang 4
	2016-11-19
发表期刊	NEUROCOMPUTING
ISSN	0925-2312
卷号	214 页码:796-804
摘要	In this paper, we present a novel appearance variation prediction model which can be embedded into the existing generative appearance model based tracking framework. Different from the existing works, which online learn appearance model with obtained tracking results, we propose to predict appearance reconstruction error. We notice that although the learned appearance model can precisely describe the target in the previous frames, the tracking result is still not accurate if in the following frame, the patch that is most similar to appearance model is assumed to be the target. We first investigate the above phenomenon by conducting experiments on two public sequences and discover that in most cases the best target is not the one with minimal reconstruction error. Then we design three kinds of features which can encode motion, appearance, appearance reconstruction error information of target's surrounding image patches, and capture potential factors that may cause variations of target's appearance as well as its reconstruction error. Finally, with these features, we learn an effective random forest for predicting reconstruction error of the target during tracking. Experiments on various datasets demonstrate that the proposed method can be combined with many existing trackers and improve their performances significantly. (C) 2016 Elsevier B.V. All rights reserved.
关键词	Object tracking Appearance model Appearance prediction
DOI	10.1016/j.neucom.2016.06.058
收录类别	SCI
语种	英语
资助项目	National Basic Research Program of China (973 Program)[2012CB316400] ; National Basic Research Program of China (973 Program)[2015CB351802] ; National Natural Science Foundation of China[61303153] ; National Natural Science Foundation of China[61332016] ; National Natural Science Foundation of China[61572465]
WOS研究方向	Computer Science
WOS类目	Computer Science, Artificial Intelligence
WOS记录号	WOS:000386741300075
出版者	ELSEVIER SCIENCE BV
引用统计	被引频次：2[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/2XEOYT63/7800
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Ma, Bingpeng
作者单位	1.Univ Chinese Acad Sci, Sch Comp & Control Engn, Beijing 100190, Peoples R China 2.Chinese Acad Sci, Inst Comp Technol, Key Lab Intelligent Informat Proc, Beijing 100080, Peoples R China 3.Chinese Acad Sci, Key Lab Big Data Min & Knowledge Management, Beijing, Peoples R China 4.Harbin Inst Technol, Sch Comp Sci & Technol, Harbin, Peoples R China
推荐引用方式 GB/T 7714	Li, Guorong,Ma, Bingpeng,Huang, Jun,et al. Beyond appearance model: Learning appearance variations for object tracking[J]. NEUROCOMPUTING,2016,214:796-804.
APA	Li, Guorong,Ma, Bingpeng,Huang, Jun,Huang, Qingming,&Zhang, Weigang.(2016).Beyond appearance model: Learning appearance variations for object tracking.NEUROCOMPUTING,214,796-804.
MLA	Li, Guorong,et al."Beyond appearance model: Learning appearance variations for object tracking".NEUROCOMPUTING 214(2016):796-804.