Embedding Motion and Structure Features for Action Recognition

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

	Embedding Motion and Structure Features for Action Recognition
	Zhen, Xiantong 1; Shao, Ling 1; Tao, Dacheng2,3 ; Li, Xuelong4
刊名	ieee transactions on circuits and systems for video technology
	2013-07-01
卷号	23 期号:7 页码:1182-1190
关键词	Biologically inspired features discriminative locality alignment human action recognition
英文摘要	we propose a novel method to model human actions by explicitly coding motion and structure features that are separately extracted from video sequences. firstly, the motion template (one feature map) is applied to encode the motion information and image planes (five feature maps) are extracted from the volume of differences of frames to capture the structure information. the gaussian pyramid and center-surround operations are performed on each of the six obtained feature maps, decomposing each feature map into a set of subband maps. biologically inspired features are then extracted by successively applying gabor filtering and max pooling on each subband map. to make a compact representation, discriminative locality alignment is employed to embed the high-dimensional features into a low-dimensional manifold space. in contrast to sparse representations based on detected interest points, which suffer from the loss of structure information, the proposed model takes into account the motion and structure information simultaneously and integrates them in a unified framework; it therefore provides an informative and compact representation of human actions. the proposed method is evaluated on the kth, the multiview ixmas, and the challenging ucf sports datasets and outperforms state-of-the-art techniques on action recognition
WOS标题词	science & technology ; technology
类目[WOS]	engineering, electrical & electronic
研究领域[WOS]	engineering
关键词[WOS]	action representation ; scene classification ; object recognition ; localization ; context ; points ; images ; cortex ; scale
收录类别	SCI ; EI
语种	英语
WOS记录号	WOS:000321276900009
公开日期	2015-06-30
内容类型	期刊论文
源URL	[http://ir.opt.ac.cn/handle/181661/24008]
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.Univ Sheffield, Dept Elect & Elect Engn, Sheffield S1 3JD, S Yorkshire, England 2.Univ Technol Sydney, Ctr Quantum Computat & Intelligent Syst, Ultimo, NSW 2007, Australia 3.Univ Technol Sydney, Fac Engn Informat Technol, Ultimo, NSW 2007, Australia 4.Chinese Acad Sci, Xian Inst Opt & Precisio Mech, State Key Lab Transient Opt & Photon, Ctr OPT IMagery Anal & Learning OPTIMAL, Xian 710119, Shaanxi, Peoples R China
推荐引用方式 GB/T 7714	Zhen, Xiantong,Shao, Ling,Tao, Dacheng,et al. Embedding Motion and Structure Features for Action Recognition[J]. ieee transactions on circuits and systems for video technology,2013,23(7):1182-1190.
APA	Zhen, Xiantong,Shao, Ling,Tao, Dacheng,&Li, Xuelong.(2013).Embedding Motion and Structure Features for Action Recognition.ieee transactions on circuits and systems for video technology,23(7),1182-1190.
MLA	Zhen, Xiantong,et al."Embedding Motion and Structure Features for Action Recognition".ieee transactions on circuits and systems for video technology 23.7(2013):1182-1190.