基于时空关联图模型的视频监控目标跟踪

张诚; 马华东; 傅慧源

doi:10.13700/j.bh.1001-5965.2014.0472

基于时空关联图模型的视频监控目标跟踪

doi: 10.13700/j.bh.1001-5965.2014.0472

北京邮电大学智能通信软件与多媒体北京市重点实验室, 北京 100876

基金项目: 国家863计划资助项目(2014AA015101); 国家自然科学基金资助项目(61402048); 国家工信部物联网发展专项资金; 北京市教育委员会共建项目

详细信息

作者简介:
张诚(1990—),男,湖北黄冈人,硕士生,zhangcheng@bupt.edu.cn

通讯作者:
马华东(1964—),男,河南南阳人,教授,mhd@bupt.edu.cn,主要研究方向为物联网与传感网、多媒体系统与网络.

中图分类号: TP391
计量
- 文章访问数: 1098
- HTML全文浏览量: 71
- PDF下载量: 837
- 被引次数: 0
出版历程
- 收稿日期: 2014-04-28
- 修回日期: 2014-08-01
- 网络出版日期: 2015-04-20

Object tracking in surveillance videos using spatial-temporal correlation graph model

Beijing Key Laboratory of Intelligent Telecommunications Software and Multimedia, Beijing University of Posts and Telecommunications, Beijing 100876, China

摘要

摘要: 多摄像机监控环境下的无重叠视域目标跟踪问题十分具有挑战性,其原因在于跟踪目标在网络中的转移与运动规律往往具有不确定性.目标跟踪的关键问题在于摄像机之间的目标关联以及如何依据网络拓扑结构来找到目标之间的对应关系.提出了一种图模型来对摄像机网络中的时空关联关系进行表达.图模型中的节点表示目标在摄像机视域中的出现区域和消失区域,边由时间与空间关系进行约束.提出了一种将目标外观模型与图模型相融合的跟踪方法,其中外观模型通过协方差描述子进行特征融合,同时,结合二部图匹配策略来解决多摄像头目标跟踪中的识别与匹配问题.在真实监控视频上的实验验证了该方法的有效性.
- 目标跟踪 /
- 多摄像机网络 /
- 时空关联图模型 /
- 二部图匹配 /
- 数据关联
Abstract: Object tracking in non-overlapping multi-camera surveillance is a challenging problem since the transition time between cameras varies greatly from individual to individual with uncertainty. The key problem of object tracking in wide areas is data association and how to find correspondences between objects via camera topology. A novel graph model was proposed to capture the spatial-temporal correlation among objects, which are moving in the camera network. Source/sink regions are graph nodes, and the graph edges are constructed by the spatial and temporal constrains. Specifically, a tracking method combining appearance model and graph model was proposed to solve the problem of object re-identification and data association via bipartite matching in multi-camera object tracking. In addition, region covariance descriptor was utilized to fuse the appearance feature. Experiments with real videos validate the proposed approach.
- object tracking /
- multi-camera network /
- spatial-temporal graph model /
- bipartite matching /
- data association

HTML全文

参考文献(20)

[1]	Wang X. Intelligent multi-camera video surveillance:a review[C]//Pattern Recognition Letters.Netherlands:Academic Press,Elsevier,2013,3(4):3-19.
[2]	Fu H,Ma H, Ming A.EGMM:an enhanced Gaussian mixture model for detecting moving objects with intermittent stops[C]//IEEE International Conference on Multimedia and Expo.Piscataway,NJ:IEEE,2011,7041(2):1-6.
[3]	Fu H,Ma H, Liu L.Robust human detection with low energy consumption in visual sensor network[C]//IEEE International Conference on Mobile Ad-hoc and Sensor Networks.Piscataway,NJ:IEEE,2011:91-97.
[4]	Zheng W S, Gong S,Xiang T.Reidentification by relative distance comparison[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2013,35(3):653-668.
[5]	Wu Y,Lim J, Yang M H.Online object tracking:a benchmark[C]//IEEE Conference on Computer Vision and Pattern Recognition.Piscataway,NJ:IEEE,2013,9(4):2411-2418.
[6]	Kalal Z, Mikolajczyk K,Matas J.Tracking-learning-detection[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2012,34(7):1409-1422.
[7]	Makris D,Ellis T, Black J.Bridging the gaps between cameras[C]//Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition.Piscataway,NJ:IEEE,2004,2:II205-II210.
[8]	Stauffer C. Estimating tracking sources and sinks[C]//IEEE Conference on Computer Vision and Pattern Recognition Workshop.Piscataway,NJ:IEEE,2003:35.
[9]	Tieu K, Dalley G,Grimson W E L.Inference of non-overlapping camera network topology by measuring statistical dependence[C]//IEEE International Conference on Computer Vision.Piscataway ,NJ:IEEE,2005,2:1842-1849.
[10]	Loy C C, Xiang T,Gong S.Multi-camera activity correlation analysis [C]//IEEE Conference on Computer Vision and Pattern Recognition.Piscataway,NJ:IEEE,2009:1988-1995.
[11]	Wang X, Tieu K,Grimson W E L.Correspondence-free activity analysis and scene modeling in multiple camera views[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2010,32(1):56-71.
[12]	刘少华,赖世铭, 张茂军.基于最小费用流模型的无重叠视域多摄像机目标关联算法[J].自动化学报,2010,36(10):1484-1489. Liu S H,Lai S M,Zhang M J.A min-cost flow based algorithm for objects association of multiple non-overlapping cameras[J].ACTA Automatica Sinica,2010,36(10):1484-1489(in Chinese).
[13]	万九卿,刘青云. 基于高阶时空模型的视觉传感网络数据关联方法[J].自动化学报,2012,38(2):236-247. Wan J Q,Liu Q Y.Data association in visual sensor networks based on high-order spatio-temporal model[J].Acta Automatica Sinica,2012,38(2):236-247(in Chinese).
[14]	Song B, Roy-Chowdhury A K.Stochastic adaptive tracking in a camera network[C]//IEEE 11th International Conference on Computer Vision.Piscataway,NJ:IEEE,2007:1-8.
[15]	Tuzel O, Porikli F,Meer P.Region covariance:a fast descriptor for detection and classification[C]//European Conference on Computer Vision.Heidelberg:Springer,2006:589-600.
[16]	Gray D,Tao H. Viewpoint invariant pedestrian recognition with an ensemble of localized features[C]//European Conference on Computer Vision.Heidelberg:Springer,2008:262-275.
[17]	Bay H, Tuytelaars T,Van Gool L.Surf:speeded up robust features[C]//European Conference on Computer Vision.Heidelberg:Springer,2006:404-417.
[18]	Förstner W, Moonen B.A metric for covariance matrices[C]//Geodesy-the Challenge of the 3rd Millennium.Heidelberg:Springer,2003:299-309.
[19]	Kuhn H W. Variants of the Hungarian method for assignment problems[J].Naval Research Logistics Quarterly,1956,3(4):253-258.
[20]	Large-scale object tracking over a multiple-camera network[EB/OL] .The First IEEE International Conference on Multimedia Big Data.[2014-04-20].http://www.bigmm2015.org/Challenge.asp.