摘要
群组行为分析是光学影像序列分析中的一项重要课题,在近年来引起了人工智能领域研究人员的广泛关注.与行人个体相比,群组能提供更高层的语义表示,为分析人群运动模式提供基础.本文将人群影像序列中的影像块作为研究对象,提出了一种基于多视角聚类的群组行为分析方法,对运动模式不同的群组进行区分,主要研究内容有:(1)提出了基于特征点的影像块构图方法,从交互关系、空间位置、运动方向分布,以及运动规律等方面衡量影像块之间的关系;(2)提出了一种多视角聚类方法,通过融合多种特征对每个影像块分配类标签,并引入图多样性正则项以避免特征冗余;(3)提出了一种类合并方法,根据类内特征点的运动方向和类中心位置坐标,对关联度较高的类别进行合并,自动确定最终群组数目.CUHK人群数据集上的实验结果证明了该方法能够准确划分出影像数据中的群组.另外,与现有方法相比,本文提出的多视角聚类方法也在不同数据集上取得了较好的实验结果.
Group behavior analysis is a hot topic in intelligent video surveillance, and has attracted a surge of interest in the field of artificial intelligence. Groups are the basic components of a crowd system, and provide a high-level representation of the crowd phenomenon. By investigating the motion dynamics within each image patch, this paper proposes a multiview-based group behavior analysis method that is able to divide the paths into different groups. The main contributions are threefold：（1） the correlation between image paths is captured from four views（interaction, distance, motion direction, and motion transition）,（2） a multiview clustering method with diversity regularization is proposed to perceive the complementary information within the multiview data and alleviate the influence of redundant features, and（3） a cluster merging strategy is designed to combine the highly correlated clusters and determine the final groups automatically. Experimental results on several benchmark datasets validate the good performance of the proposed method.
作者
李学龙
陈穆林
王琦
Xuelong LI;Mulin CHEN;Qi WANG(Xi＇an Institute of Optics and Precision Mechanics,Chinese Academy of Sciences,Xi＇an 710119,China;School of Computer Science and Center for OPTical IMagery Analysis and Learning（OPTIMAL）,Northwester~ Polytechnical University,Xi＇an 710072,China;Unmanned System Research Institute（USRI）,Northwestern Polytechnical University,Xi＇an 710072,China)
出处
《中国科学：信息科学》
CSCD
北大核心
2018年第9期1227-1241,共15页
Scientia Sinica(Informationis)
关键词
人群分析
群组行为分析
聚类算法
图聚类
多视角聚类
crowd analysis
group behavior analysis
clustering
graph clustering
multiview clustering
