[1]康永欣,袁丽英. 人群密集场景中运动模式识别[J].哈尔滨理工大学学报,2019,(03):74-81.[doi:10.15938/j.jhust.2019.03.012]
 KANG Yong xin,YUAN Li ying. Motion Patterns Learning in Crowded Scenes[J].哈尔滨理工大学学报,2019,(03):74-81.[doi:10.15938/j.jhust.2019.03.012]
点击复制

 人群密集场景中运动模式识别()
分享到:

《哈尔滨理工大学学报》[ISSN:1007-2683/CN:23-1404/N]

卷:
期数:
2019年03期
页码:
74-81
栏目:
计算机与控制工程
出版日期:
2019-06-24

文章信息/Info

Title:
 Motion Patterns Learning in Crowded Scenes
文章编号:
1007-2683(2019)03-0074-08
作者:
 康永欣袁丽英
 (哈尔滨理工大学 自动化学院,黑龙江 哈尔滨 150080)
Author(s):
 KANG YongxinYUAN Liying
 (School of Automation, Harbin University of Science and Technology, Harbin 150080, China)
关键词:
 运动模式识别人群密集场景非参贝叶斯模型吉布斯采样
Keywords:
 Motion pattern learning Crowded scenes Nonparametric Bayesian model Gibbs sampling
分类号:
TP391.4
DOI:
10.15938/j.jhust.2019.03.012
文献标志码:
A
摘要:
 针对从人群密集场景中识别运动模式的问题,提出了距离依赖中餐馆连锁店过程混合模型。该模型是一种引入依赖关系的层次化非参贝叶斯模型,能够通过引进独立于数据观测值的距离依赖信息,准确建模自然分组的数据,从中挖掘共享的数据模式。给出了模型的建立过程,并通过吉布斯采样的方法对模型进行求解,同时展示了相关的实验结果。通过对纽约广场火车站监控场景数据集中47 866条片段轨迹的建模分析,证明了模型可以自动确定场景中运动模式的个数,从不完整的轨迹中以98%的正确度学习并表达运动模式,并且能够在不同的运动模式之间共享公共的子模式。
Abstract:
 To address the motion patterns learning task in crowded scenes, we propose a novel Distance Dependent Chinese Restaurant Franchise (DDCRF) mixture model, which is a hierarchical nonparametric Bayesian model based on dependencies DDCRF can learn the latent patterns accurately by introducing the distance information which is dependent of the observations of data points. We detail the generative process and Gibbs sampling process of DDCRF, and then the results of experiments are shown.  An extensive evaluation is performed on the dataset including 47,866 tracklets collected from the crowded New York Grand Central station, indicating that our algorithm has the following advantages: deducing the number of latent motion patterns automatically, learning motion patterns precisely from these tracklets, and sharing constituent parts and sub  patterns among different motion patterns.

参考文献/References:

[1]朱丽英,梁辰,张鸿洲. 基于运动目标的监控视频检索方法[J]. 计算机应用与软件,2011(12):96.
[2]吕英丽,顾勇,张晓峰. 基于HMM监控视频的异常事件检测[J].数据采集与处理,2014(6):1030.
[3]KWON Y, KANG K, JIN J,et al. Hierarchically Linked Infinite Hiddenmarkov Model Based Trajectory Analysis and Semantic Region Retrieval in a Trajectory Dataset. Expert Systems with Applications, 2017(78):386.
[4]SHAO Z, LI Y. Integral Invariants for Space Motion Trajectory Matching and Recognition. Pattern Recognition, 2015,48(8):2418.
[5]MORRIS B T, TRIVEDI M M. A Survey of Visionbased Trajectory Learning and Analysis for Surveillance. IEEE transactions on circuits and systems for video technology, 2008,18(8):1114.
[6]WANG X, MA X, GRIMSON W E L. Unsupervised Activity Perception in Crowded and Complicated Scenes Using Hierarchical Bayesian Models. IEEE Transactions on pattern analysis and machine intelligence, 2009,31(3):539.
[7]HU W, XIAO X, FU Z, et al. A System for Learning Statistical Motion Patterns. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2006,28(9):1450.
[8]ERGEZER H, LEBLEBICIOGLU K. Anomaly Detection and Activity Perception Using Covariance Descriptor for Trajectories. In European Conference on Computer Vision, pages 2016:728.
[9]CHONGJING W, XU Z, YI Z, et al. Analyzing Motion Patterns in Crowded Scenes Via Automatic Tracklets Clustering[J]. china communications, 2013, 10(4): 144.
[10]SHAO J, CHANGE Loy C, WANG X. Sceneindependent Group Profiling in Crowd[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2014: 2219.
[11]ZHAO W, ZHANG Z, HUANG K. Joint Crowd Detection and Semantic Scene Modeling Using a Gestalt Lawsbased Similarity[C]//Image Processing (ICIP), 2016 IEEE International Conference on. IEEE, 2016: 1220.
[12]ZHOU B, WANG X, TANG X.Random Field Topic Model for Semantic Region Analysis in Crowded Scenes from Tracklets[C]//Computer Vision and Pattern Recognition (CVPR), 2011 IEEE Conference on. IEEE, 2011: 3441. 
[13]BLEI D M, NG A Y, JORDAN M I. Latent Dirichlet Allocation[J]. Journal of machine Learning research, 2003, 3(Jan): 993.
[14]BLEI D M, FRAZIER P I. Distance Dependent Chinese Restaurant Processes[J]. Journal of Machine Learning Research, 2011, 12(Aug): 2461.
[15]FERGUSON T S. A Bayesian Analysis of Some Nonparametric Problems[J]. The annals of statistics, 1973: 209.
[16]TEH Y W, JORDAN M I, BEAL M J, et al. Sharing Clusters Among Related Groups: Hierarchical Dirichlet Processes[C]//Advances in neural information processing systems. 2005: 1385.
[17]EMONET R, VARADARAJAN J, ODOBEZ J M. Extracting and Locating Temporal Motifs in Video Scenes Using a Hierarchical Non Parametric Bayesian Model[C]//Computer Vision and Pattern Recognition (CVPR), 2011 IEEE Conference on. IEEE, 2011: 3233.
[18]KUETTEL D, BREITENSTEIN M D, VAN Gool L, et al. What′s Going on Discovering Spatiotemporal Dependencies in Dynamic Scenes[C]//Computer Vision and Pattern 
[19]Recognition (CVPR), 2010 IEEE Conference on. IEEE, 2010: 1951.
[20]WANG X, MA X,GRIMSON W E L. Unsupervised Activity Perception in Crowded and Complicated Scenes Using Hierarchical Bayesian Models[J]. IEEE Transactions on pattern analysis and machine intelligence, 2009, 31(3): 539.
[21]MOBERTS B, VILANOVA A, VAN Wijk J J. Evaluation of Fiber Clustering Methods for Diffusion Tensor Imaging[C]//Visualization, 2005. VIS 05. IEEE. IEEE, 2005: 65.
[22]WANG X, MA K T, NG G W, et al. Trajectory Analysis and Semantic Region Modeling Using a Nonparametricbayesian Model[C]//Computer Vision and Pattern Recognition, 2008. CVPR 2008. IEEE Conference on. IEEE, 2008: 1.
[23]ATEV S, MASOUD O, PAPANIKOLOPOULOS N. Learning Traffic Patterns at Intersections by Spectral Clustering of Motion Trajectories. In Intelligent Robots and Systems[C]// 2006 IEEE/RSJ International Conference on, IEEE, 2006:4851.

相似文献/References:

[1]孙永全,郭建英,陈洪科,等.AMSAA模型可靠性增长预测方法的改进[J].哈尔滨理工大学学报,2010,(05):49.
 SUN Yong-quan,GUO Jian-ying,CHEN Hong-ke,et al.An Improved Reliability Growth Prediction Algorithm Based on AMSAA Model[J].哈尔滨理工大学学报,2010,(03):49.
[2]滕志军,李晓霞,郑权龙,等.矿井巷道的MIMO信道几何模型及其信道容量分析[J].哈尔滨理工大学学报,2012,(02):14.
 TENG Zhi-jun,LI Xiao-xia,ZHENG Quan-long.Geometric Model for Mine MIMO Channels and Its Capacity Analysis[J].哈尔滨理工大学学报,2012,(03):14.
[3]李艳苹,张礼勇.新训练序列下的改进OFDM符号定时算法[J].哈尔滨理工大学学报,2012,(02):19.
 LI Yan-ping,ZHANG Li-yong.An Improved Algorithm of OFDM Symbol Timing Based on A New Training Sequence[J].哈尔滨理工大学学报,2012,(03):19.
[4]赵彦玲,车春雨,铉佳平,等.钢球全表面螺旋线展开机构运动特性分析[J].哈尔滨理工大学学报,2013,(01):37.
 ZHAO Yan-ling,CHE Chun-yu,XUAN Jia-ping,et al.[J].哈尔滨理工大学学报,2013,(03):37.
[5]李冬梅,卢旸,刘伟华,等.一类具有连续接种的自治SEIR传染病模型[J].哈尔滨理工大学学报,2013,(01):73.
 LI Dong-mei,LU Yang,LIU Wei-hua.[J].哈尔滨理工大学学报,2013,(03):73.
[6]华秀英,刘文德.奇Hamiltonian李超代数偶部的非负Z-齐次导子空间[J].哈尔滨理工大学学报,2013,(01):76.
 HUA Xiu-ying,LIU Wen-de.[J].哈尔滨理工大学学报,2013,(03):76.
[7]桂存兵,刘洋,何业军,等.基于LCC谐振电路阻抗匹配的光伏发电最大功率点跟踪[J].哈尔滨理工大学学报,2013,(01):90.
 GUI Cun-bing,LIU Yong,HE Ye-jun.[J].哈尔滨理工大学学报,2013,(03):90.
[8]翁凌,闫利文,夏乾善,等.PI/TiC@Al2O3复合薄膜的制备及其电性能研究[J].哈尔滨理工大学学报,2013,(02):25.
 WENG Ling,YAN Li-wen,XIA Qian-shan.[J].哈尔滨理工大学学报,2013,(03):25.
[9]姜彬,林爱琴,王松涛,等.高速铣刀安全性设计理论与方法[J].哈尔滨理工大学学报,2013,(02):63.
 JIANG Bin,LIN Ai-qin,WANG Song-tao,et al.[J].哈尔滨理工大学学报,2013,(03):63.
[10]李星纬,李晓东,张颖彧,等.EVOH 磺酸锂电池隔膜的制备及微观形貌[J].哈尔滨理工大学学报,2013,(05):18.
 LI Xing- wei,LI Xiao- dong,ZHANG Ying- yu,et al.The Preparation and Microcosmic Morphology oEVOH- SO Li Lithium Ion Battery Septum[J].哈尔滨理工大学学报,2013,(03):18.

备注/Memo

备注/Memo:
 收稿日期: 2017-07-12
基金项目: 国家自然科学基金(61305001)
作者简介:
袁丽英(1971—),女,教授,硕士研究生导师
通信作者:
康永欣(1991—),女,硕士研究生,E-mail:kangyongxin2015@iaac.cn
更新日期/Last Update: 2019-06-20