Learning Relevance Restricted Boltzmann Machine for Unstructured Group Activity and Event Understanding

doi:10.1007/s11263-016-0896-3

CORC > 自动化研究所 > 中国科学院自动化研究所 > 智能感知与计算研究中心

	Learning Relevance Restricted Boltzmann Machine for Unstructured Group Activity and Event Understanding
	Zhao, Fang 1; Huang, Yongzhen1,3 ; Wang, Liang1,3 ; Xiang, Tao 2; Tan, Tieniu1
刊名	INTERNATIONAL JOURNAL OF COMPUTER VISION
	2016-09-01
卷号	119 期号:3 页码:329-345
关键词	Representation Learning Video Analysis Restricted Boltzmann Machine Sparse Bayesian Learning
DOI	10.1007/s11263-016-0896-3
文献子类	Article
英文摘要	Analyzing unstructured group activities and events in uncontrolled web videos is a challenging task due to (1) the semantic gap between class labels and low-level visual features, (2) the demanding computational cost given high-dimensional low-level feature vectors and (3) the lack of labeled training data. These difficulties can be overcome by learning a meaningful and compact mid-level video representation. To this end, in this paper a novel supervised probabilistic graphical model termed Relevance Restricted Boltzmann Machine (ReRBM) is developed to learn a low-dimensional latent semantic representation for complex activities and events. Our model is a variant of the Restricted Boltzmann Machine (RBM) with a number of critical extensions: (1) sparse Bayesian learning is incorporated into the RBM to learn features which are relevant to video classes, i.e., discriminative; (2) binary stochastic hidden units in the RBM are replaced by rectified linear units in order to better explain complex video contents and make variational inference tractable for the proposed model; and (3) an efficient variational EM algorithm is formulated for model parameter estimation and inference. We conduct extensive experiments on two recent challenging benchmarks: the Unstructured Social Activity Attribute dataset and the Event Video dataset. The experimental results demonstrate that the relevant features learned by our model provide better semantic and discriminative description for videos than a number of alternative supervised latent variable models, and achieves state of the art performance in terms of classification accuracy and retrieval precision, particularly when only a few labeled training samples are available.
WOS关键词	LATENT DIRICHLET ALLOCATION ; CLASSIFICATION ; ALGORITHM ; MODELS
WOS研究方向	Computer Science
语种	英语
WOS记录号	WOS:000380270000008
资助机构	National Basic Research Program of China(2012CB316300) ; National Natural Science Foundation of China(61525306 ; Strategic Priority Research Program of the CAS(XDB02070100) ; 61573354 ; 61135002 ; 61420106015)
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/12170]
专题	自动化研究所_智能感知与计算研究中心
作者单位	1.Chinese Acad Sci, Inst Automat, Ctr Res Intelligent Percept & Comp, Beijing, Peoples R China 2.Queen Mary Univ London, Sch Elect Engn & Comp Sci, London, England 3.Chinese Acad Sci, Ctr Excellence Brain Sci & Intelligence Technol, Beijing, Peoples R China
推荐引用方式 GB/T 7714	Zhao, Fang,Huang, Yongzhen,Wang, Liang,et al. Learning Relevance Restricted Boltzmann Machine for Unstructured Group Activity and Event Understanding[J]. INTERNATIONAL JOURNAL OF COMPUTER VISION,2016,119(3):329-345.
APA	Zhao, Fang,Huang, Yongzhen,Wang, Liang,Xiang, Tao,&Tan, Tieniu.(2016).Learning Relevance Restricted Boltzmann Machine for Unstructured Group Activity and Event Understanding.INTERNATIONAL JOURNAL OF COMPUTER VISION,119(3),329-345.
MLA	Zhao, Fang,et al."Learning Relevance Restricted Boltzmann Machine for Unstructured Group Activity and Event Understanding".INTERNATIONAL JOURNAL OF COMPUTER VISION 119.3(2016):329-345.