Realistic action recognition via sparsely-constructed Gaussian processes

CORC > 西安光学精密机械研究所 > 中国科学院西安光学精密机械研究所 > 光学影像学习与分析中心

	Realistic action recognition via sparsely-constructed Gaussian processes
	Liu, Li 1,2; Shao, Ling 1,2; Zheng, Feng 2; Li, Xuelong3
刊名	pattern recognition
	2014-12-01
卷号	47 期号:12 页码:3819-3827
关键词	Action recognition Gaussian processes l(1) construction Local approximation
ISSN号	0031-3203
产权排序	3
合作状况	国际
英文摘要	realistic action recognition has been one of the most challenging research topics in computer vision. the existing methods are commonly based on non-probabilistic classification, predicting category labels but not providing an estimation of uncertainty. in this paper, we propose a probabilistic framework using gaussian processes (gps), which can tackle regression problems with explicit uncertain models, for action recognition. a major challenge for gps when applied to large-scale realistic data is that a large covariance matrix needs to be inverted during inference. additionally, from the manifold perspective, the intrinsic structure of the data space is only constrained by a local neighborhood and data relationships with far-distance usually can be ignored. thus, we design our gps covariance matrix via the proposed l(1) construction and a local approximation (la) covariance weight updating method, which are demonstrated to be robust to data noise, automatically sparse and adaptive to the neighborhood. extensive experiments on four realistic datasets, i.e., ucf youtube, ucf sports, hollywood2 and hmdb51, show the competitive results of l(1)-gps compared with state-of-the-art methods on action recognition tasks. (c) 2014 elsevier ltd. all rights reserved.
WOS标题词	science & technology ; technology
类目[WOS]	computer science, artificial intelligence ; engineering, electrical & electronic
研究领域[WOS]	computer science ; engineering
收录类别	SCI ; EI
语种	英语
WOS记录号	WOS:000342870900007
公开日期	2015-03-19
内容类型	期刊论文
源URL	[http://ir.opt.ac.cn/handle/181661/22419]
专题	西安光学精密机械研究所_光学影像学习与分析中心
作者单位	1.Nanjing Univ Informat Sci & Technol, Coll Elect & Informat Engn, Nanjing 210044, Jiangsu, Peoples R China 2.Univ Sheffield, Dept Elect & Elect Engn, Sheffield S1 3JD, S Yorkshire, England 3.Chinese Acad Sci, XIOPM, State Key Lab Transient Opt & Photon, Xian 710119, Peoples R China
推荐引用方式 GB/T 7714	Liu, Li,Shao, Ling,Zheng, Feng,et al. Realistic action recognition via sparsely-constructed Gaussian processes[J]. pattern recognition,2014,47(12):3819-3827.
APA	Liu, Li,Shao, Ling,Zheng, Feng,&Li, Xuelong.(2014).Realistic action recognition via sparsely-constructed Gaussian processes.pattern recognition,47(12),3819-3827.
MLA	Liu, Li,et al."Realistic action recognition via sparsely-constructed Gaussian processes".pattern recognition 47.12(2014):3819-3827.