动作识别与行为理解综述  

Action Recognition and Activity Understanding:A Review

作　　者：徐光祐[1] 曹媛媛[1] 

机构地区：[1]清华大学计算机科学与技术系普适计算教育部重点实验室,北京100084

出　　处：《中国图象图形学报》2009年第2期189-195,共7页Journal of Image and Graphics

基　　金：国家自然科学基金项目(60673189)

摘　　要：随着"以人为中心计算"的兴起和生活中不断涌现的新应用,动作识别和行为理解逐渐成为计算机视觉领域的研究热点。主要从视觉处理的角度分析了动作识别和行为理解的研究现状,从行为的定义、运动特征提取和动作表示以及行为理解的推理方法3个方面对目前的工作做了分析和比较,并且指出了目前这些工作面临的难题和今后的研究方向。As the ＂ Human-centered computing＂ is getting more popular and novel applications are evolving, actionrecognition and activity understanding are attracting researchers in the field of computer vision. In this paper, we review the state-of-the-art work on action and activity analysis with focus on three parts： Definition of activity, low-level motion features extraction and action representation, and reasoning method for activity understanding. Furthermore, open problems for future research and potential directions are discussed

关 键 词：以人为中心计算 动作识别 行为理解 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]