Group sparse representation for image categorization and semantic video retrieval  

作　　者：LIU YaNan WU Fei ZHUANG YueTing 

机构地区：[1]School of Information, Zhejiang University of Finance & Economics, Hangzhou, 310018, China [2]College of Computer Science and Technology, Zhejiang University, Hangzhou, 310012, China

出　　处：《Science China(Information Sciences)》2011年第10期2051-2063,共13页中国科学（信息科学）（英文版）

基　　金：supported by the National Basic Research Program of China (Grant No.2010CB327905);the National Natural Science Foundation of China (Grant Nos.60833006,90920303);the National High Technology Research and Development Program (Grant No.2006AA010107);the National Key Technology R&D Program(Grant No.2007BAH11B06);the Natural Science Foundation of Zhejiang Province (Grant No.Y1100773)

摘　　要：Multimedia content analysis and management are a promising and challenging theme. In this paper we develop a novel approach to image representation, which we call group sparse representation （GSR）, for image classification and video retrieval. The basic idea is to represent a test image as a weighted combination of all the training images. In particular, we introduce two sets of weight coefficients, one for each training image and the other for each class. Moreover, we formulate our concern as a group nonnegative garrote model. The resulting representations are sparse, and they are appropriate for discriminant analysis. Experiments on Caltech101 and PASCAL VOC2008 image dataset and TRECVID2005 video corpus testify that our proposed approach is efficient and effective.Multimedia content analysis and management are a promising and challenging theme. In this paper we develop a novel approach to image representation, which we call group sparse representation （GSR）, for image classification and video retrieval. The basic idea is to represent a test image as a weighted combination of all the training images. In particular, we introduce two sets of weight coefficients, one for each training image and the other for each class. Moreover, we formulate our concern as a group nonnegative garrote model. The resulting representations are sparse, and they are appropriate for discriminant analysis. Experiments on Caltech101 and PASCAL VOC2008 image dataset and TRECVID2005 video corpus testify that our proposed approach is efficient and effective.

关 键 词：compressive sensing group sparse representation bag-of-SIFT-words nonnegative garrote llminimization apcluster 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术] TN911.73[自动化与计算机技术—计算机科学与技术]