Complex event recognition using constrained low-rank representation

作     者：Dehghan, Afshin Oreifej, Omar Shah, Mubarak 

作者机构：Univ Cent Florida Orlando FL 32816 USA Univ Calif Berkeley Berkeley CA 94720 USA 

出 版 物：《IMAGE AND VISION COMPUTING》 (图像与视觉计算)

年 卷 期：2015年第42卷

页      面：13-21页

核心收录：

学科分类：0808[工学-电气工程] 08[工学] 0835[工学-软件工程] 0812[工学-计算机科学与技术（可授工学、理学学位）] 0702[理学-物理学] 

基　　金：Intelligence Advanced Research Projects Activity (IARPA) via the Department of Interior National Business Center [D11PC20066] 

主　　题：Complex event recognition Low-rank optimization Activity recognition Action concepts 

摘      要：Complex event recognition is the problem of recognizing events in long and unconstrained videos. In this extremely challenging task, concepts have recently shown a promising direction where core low-level events (referred to as concepts) are annotated and modeled using a portion of the training data, then each complex event is described using concept scores, which are features representing the occurrence confidence for the concepts in the event. However, because of the complex nature of the videos, both the concept models and the corresponding concept scores are significantly noisy. In order to address this problem, we propose a novel low-rank formulation, which combines the precisely annotated videos used to train the concepts, with the rich concept scores. Our approach finds a new representation for each event, which is not only low-rank, but also constrained to adhere to the concept annotation, thus suppressing the noise, and maintaining a consistent occurrence of the concepts in each event. Extensive experiments on large scale real world dataset TRECVID Multimedia Event Detection 2011 and 2012 demonstrate that our approach consistently improves the discriminativity of the concept scores by a significant margin. (C) 2015 Published by Elsevier B.V.