视频行人检测及跟踪的关键技术研究
|
摘要
目标检测和视觉跟踪是计算机视觉领域的基本内容和研究热点，在视频监控、智能交通、机器人导航、智能车辆驾驶辅助和人机交互等领域有着重要的应用。实际公共场所中遍布着各种摄像头来辅助管理和安全监控，如地铁、机场、银行和交通路口等地。这些应用主要关注的对象是行人和汽车等，非常依赖于目标检测和视觉跟踪方法。目标检测和视觉跟踪二者关系密切并且都存在不少问题。这两个方向关键技术的研究发展及其在视频监控和智能交通等领域的应用有着重要的意义。本文面向视频监控和智能交通，对视觉跟踪、多人跟踪和行人检测的一些关键问题展开了研究和改进。
视觉跟踪着重在跟踪的实用性上，即面对常发生的局部遮挡、外观变化、转动、背景干扰等困难情况，能够持续稳健地跟踪目标。单一特征或模型能够提取或学习目标的部分视觉特性，具有一些优点且能较好地解决一些跟踪问题。然而对于视觉跟踪，单一特征或模型也常有其缺点和不足之处。本文对于两种流行的视觉跟踪方法结合其他的特征或模型进行了研究改进，使得这两种方法提高了跟踪性能和稳健性。本文所做研究工作和创新性贡献主要表现如下：
(1)提出了一种基于合成子空间的跟踪方法。增量本征子空间能有效地学习目标外观的主要模式且跟踪性能好。然而不能适应较快的外观变化。线性子空间恰能迅速学习最新的外观模式，本文将二者巧妙地结合起来得到合成子空间，同时具备二者的优点，提高了跟踪方法的适应能力和稳健性。
(2)提出了一种基于特征选择的多子块跟踪方法。多子块跟踪方法效率高且能解决遮挡问题。子块生成方法和跟踪特征是其关键技术。RGB加权特征能克服子块直方图的稀疏问题，并可选择出当前的最优区分特征来适应目标外观和背景的变化。基于最优特征，提出了一种生成跟踪子块的方法。该跟踪方法能在遮挡和背景复杂等情况下稳健地跟踪一般性目标。
近几年，多行人跟踪和行人检测发展迅速，也是视频监控等领域的关键技术。本文所做研究工作和创新性贡献主要表现如下：
(1)提出了一种基于在线双层关联的多行人跟踪方法。基于子轨迹的多行人跟踪方法是近年的一类优秀方法。该类方法包含生成子轨迹和关联子轨迹两个部分。针对串联检测结果生成子轨迹的方法，采用粒子滤波跟踪来进行改进。有效地改善了因漏检和虚警等导致的子轨迹过分散问题，获得更准确及更长的子轨迹。先前方法在很长时间后才对子轨迹进行全局关联。本文提出一种在线关联子轨迹的方法，能实现准实时跟踪关联。实验结果表明此方法跟踪性能好，在实际拥挤场景下能有效地进行多行人跟踪。
(2)针对视频监控，提出了一种基于Tff处理的视频行人检测方法。尽管目前行人检测已取得显著进展，针对监控场景研究运动特征的检测工作却很少。提出一种Tff处理，通过帧间的灰度变化提取运动信息。提出Tff幅值特征来训练预检器，能迅速地排除大量背景。提出区分性强的Tff方向直方图特征来训练行人检测器。该方法检测效率高且性能好，可用于视频监控或智能交通中。
Object detection and visual tracking are basic and hot research topics in the field ofcomputer vision, and serve as indispensable aspects in video surveillance, robot navigation,intelligent traffic, intelligent driving assistance, human-computer interaction etc. It is knownto all that cameras are prevalently used in public places for better management, surveillanceand safty purposes, such as subway stations, banks, airports and traffic intersections. Suchapplications focus on pedestrians and cars, and are seriously dependent on object detectionand visual tracking technology. These two topics are in close relationship, and there are stillmany problems for themselves. It is meaningful and valuable to do research on these topicsand develop application technology for video surveillance and intelligent traffic. For videosurveillance and intelligent traffic, we do research on key issues of visual tracking, multiple
people tracking and pedestrian detection.The goal of visual tracking is to robustly track objects under common difficulties, such asocclusion, changing appearance, turning and background clutter. Single feature or model cancapture or learn some aspects of the whole visual characteristic of the targets, and resolvesome tracking difficulties. However, features and models are also with disadvantages forvisual tracking. We do research on two good tracking methods, and make improvements withother features or models. Some improvements and innovations are listed as follows.
(1) A composite subspace based tracking method is proposed. Incremental principalcomponent analysis learns the principal pattern of object appearance effectively,leading to good tracking ability. However, it is difficult to capture the current patternwhen the appearance is changing rapidly. There is a linear subspace method whichcan learn latest appearance. We combine them artfully, resulting in a compositesubspace with merits of both. The composite subspace tracking method track robustly with high adaptability.
(2) A multiple-patch tracking method with feature selection is proposed. Multiple-patchtracking is an effective method, and can deal with occlusion problem. The patchesand features are the keys of the method. And current best discriminative features canbe selected to improve the tracking performance. A patch generating method based onweighting features is also proposed. The proposed tracking method has goodperformance, and can track robustly under occlusion and complex backgrounddifficulties.
Multiple people tracking and pedestrian detection have developed rapidly in recent years.They are the key technologies to video surveillance. Some improvements and innovations arelisted as follows.
(1) A multiple people tracking method based on online two-stage association is proposed.Multiple people tracking based on tracklet is one of the best methods recently. Wepropose to improve tracklets by combining partical filtering tracking method insteadof only connecting detection responses. Scattering problem caused by missing andfalse detection is improved effectively, and long tracklets are obtained. Previousmethods make tracklet association after a long time. We propose a method whichassociates tracklets after each short interval, and enables near real-time peopleassociation. The experimental results indicate that the method obtains goodperformance, and can be applied in crowded scene.
(2) For video surveillance, a Two-frame-filtering(Tff) processing based video pedestriandetection method is proposed. Although pedestrian detection has achieved prominentdevelopment in recent years, few works research detection using motion cue forsurveillance. Tff processing which exploit motion cue by the gray value variationbetween two frames is proposed. Tff gradient feature are proposed and used to train apre-detector to exclude most of the background regions. Histogram of Tff orientedgradient(HTffOG) is also proposed. The discriminative HTffOG is used to trainpedestrian detector. Experimental results show that this effective method achievesgood performance, and is suitable for real-time surveillance and intelligent trafficapplications.
视觉跟踪着重在跟踪的实用性上，即面对常发生的局部遮挡、外观变化、转动、背景干扰等困难情况，能够持续稳健地跟踪目标。单一特征或模型能够提取或学习目标的部分视觉特性，具有一些优点且能较好地解决一些跟踪问题。然而对于视觉跟踪，单一特征或模型也常有其缺点和不足之处。本文对于两种流行的视觉跟踪方法结合其他的特征或模型进行了研究改进，使得这两种方法提高了跟踪性能和稳健性。本文所做研究工作和创新性贡献主要表现如下：
(1)提出了一种基于合成子空间的跟踪方法。增量本征子空间能有效地学习目标外观的主要模式且跟踪性能好。然而不能适应较快的外观变化。线性子空间恰能迅速学习最新的外观模式，本文将二者巧妙地结合起来得到合成子空间，同时具备二者的优点，提高了跟踪方法的适应能力和稳健性。
(2)提出了一种基于特征选择的多子块跟踪方法。多子块跟踪方法效率高且能解决遮挡问题。子块生成方法和跟踪特征是其关键技术。RGB加权特征能克服子块直方图的稀疏问题，并可选择出当前的最优区分特征来适应目标外观和背景的变化。基于最优特征，提出了一种生成跟踪子块的方法。该跟踪方法能在遮挡和背景复杂等情况下稳健地跟踪一般性目标。
近几年，多行人跟踪和行人检测发展迅速，也是视频监控等领域的关键技术。本文所做研究工作和创新性贡献主要表现如下：
(1)提出了一种基于在线双层关联的多行人跟踪方法。基于子轨迹的多行人跟踪方法是近年的一类优秀方法。该类方法包含生成子轨迹和关联子轨迹两个部分。针对串联检测结果生成子轨迹的方法，采用粒子滤波跟踪来进行改进。有效地改善了因漏检和虚警等导致的子轨迹过分散问题，获得更准确及更长的子轨迹。先前方法在很长时间后才对子轨迹进行全局关联。本文提出一种在线关联子轨迹的方法，能实现准实时跟踪关联。实验结果表明此方法跟踪性能好，在实际拥挤场景下能有效地进行多行人跟踪。
(2)针对视频监控，提出了一种基于Tff处理的视频行人检测方法。尽管目前行人检测已取得显著进展，针对监控场景研究运动特征的检测工作却很少。提出一种Tff处理，通过帧间的灰度变化提取运动信息。提出Tff幅值特征来训练预检器，能迅速地排除大量背景。提出区分性强的Tff方向直方图特征来训练行人检测器。该方法检测效率高且性能好，可用于视频监控或智能交通中。
Object detection and visual tracking are basic and hot research topics in the field ofcomputer vision, and serve as indispensable aspects in video surveillance, robot navigation,intelligent traffic, intelligent driving assistance, human-computer interaction etc. It is knownto all that cameras are prevalently used in public places for better management, surveillanceand safty purposes, such as subway stations, banks, airports and traffic intersections. Suchapplications focus on pedestrians and cars, and are seriously dependent on object detectionand visual tracking technology. These two topics are in close relationship, and there are stillmany problems for themselves. It is meaningful and valuable to do research on these topicsand develop application technology for video surveillance and intelligent traffic. For videosurveillance and intelligent traffic, we do research on key issues of visual tracking, multiple
people tracking and pedestrian detection.The goal of visual tracking is to robustly track objects under common difficulties, such asocclusion, changing appearance, turning and background clutter. Single feature or model cancapture or learn some aspects of the whole visual characteristic of the targets, and resolvesome tracking difficulties. However, features and models are also with disadvantages forvisual tracking. We do research on two good tracking methods, and make improvements withother features or models. Some improvements and innovations are listed as follows.
(1) A composite subspace based tracking method is proposed. Incremental principalcomponent analysis learns the principal pattern of object appearance effectively,leading to good tracking ability. However, it is difficult to capture the current patternwhen the appearance is changing rapidly. There is a linear subspace method whichcan learn latest appearance. We combine them artfully, resulting in a compositesubspace with merits of both. The composite subspace tracking method track robustly with high adaptability.
(2) A multiple-patch tracking method with feature selection is proposed. Multiple-patchtracking is an effective method, and can deal with occlusion problem. The patchesand features are the keys of the method. And current best discriminative features canbe selected to improve the tracking performance. A patch generating method based onweighting features is also proposed. The proposed tracking method has goodperformance, and can track robustly under occlusion and complex backgrounddifficulties.
Multiple people tracking and pedestrian detection have developed rapidly in recent years.They are the key technologies to video surveillance. Some improvements and innovations arelisted as follows.
(1) A multiple people tracking method based on online two-stage association is proposed.Multiple people tracking based on tracklet is one of the best methods recently. Wepropose to improve tracklets by combining partical filtering tracking method insteadof only connecting detection responses. Scattering problem caused by missing andfalse detection is improved effectively, and long tracklets are obtained. Previousmethods make tracklet association after a long time. We propose a method whichassociates tracklets after each short interval, and enables near real-time peopleassociation. The experimental results indicate that the method obtains goodperformance, and can be applied in crowded scene.
(2) For video surveillance, a Two-frame-filtering(Tff) processing based video pedestriandetection method is proposed. Although pedestrian detection has achieved prominentdevelopment in recent years, few works research detection using motion cue forsurveillance. Tff processing which exploit motion cue by the gray value variationbetween two frames is proposed. Tff gradient feature are proposed and used to train apre-detector to exclude most of the background regions. Histogram of Tff orientedgradient(HTffOG) is also proposed. The discriminative HTffOG is used to trainpedestrian detector. Experimental results show that this effective method achievesgood performance, and is suitable for real-time surveillance and intelligent trafficapplications.
引文
[Ada06] Adam A., Rivlin E., Shimshoni I. Robust fragments-based tracking using the integralhistogram. Proceedings of IEEE Conference on Computer Vision and Pattern Recognition,2006,1,798-805.
[Aga02] Agarwal S., Roth D. Learning a sparse representation for object detection.Proceedings of European Conference on Computer Vision,2002,4,113-127.
[And09] Andriluka M., Roth S., Schiele B. Pictorial structures revisited: People detection andarticulated pose estimation. Proceedings of IEEE Conference on Computer Vision and PatternRecognition,2009,1014-1021.
[Bab09] Babenko B., Yang M. H., Belongie S. Visual tracking with online Multiple InstanceLearning. Proceedings of IEEE Conference on Computer Vision and Pattern Recognition,2009,983-990.
[Bay06] Bay H., Tuytelaars T., Gool L. J. Surf: Speeded up robust features. Proceedings ofEuropean Conference on Computer Vision,2006,1,404-417.
[Bel09] Beleznai C., Bischof H. Fast human detection in crowded scenes by contourintegration and local shape estimation. Proceedings of IEEE Conference on Computer Visionand Pattern Recognition,2009,2246-2253.
[Ben11a] Benfold B., Reid I. D. Stable Multi-Target Tracking in Real-Time SurveillanceVideo. Proceedings of IEEE Conference on Computer Vision and Pattern Recognition,2011,3457-3464.
[Ben11b] Benenson R., Timofte R., Gool v. L. Stixels estimation without depthmapcomputation. Proceedings of the Computer Vision in Vehicle Technology Workshop,2011.
[Ben12] Benenson R., Mathias M., Timofte R., etc. Pedestrian detection at100frames persecond.Proceedings of IEEE Conference on Computer Vision and Pattern Recognition,2012,2903-2910.
[Ber06] Berclaz J., Fleuret F., Fua P. Robust people tracking with global trajectoryoptimization. Proceedings of IEEE International Conference on Computer Vision,2006,1,744-750.
[Ber08] Bernardin K., Stiefelhagen R. Evaluating multipe object tracking performance: theCLEARMOT metrics. EURASIP Journal on Image and Video Processing,2008.
[Bla98] Black M. J., Jepson A. D. EigenTracking: Robust matching and tracking ofarticulated objects using a view-based representation. International Journal of ComputerVision,1998,26,63-84.
[Bre09] Breitenstein M. D., Reichlin F., Leibe B., etc. Robust tracking-by-detection using adetector confidence particle filter. Proceedings of IEEE International Conference onComputer Vision,2009,1515-1522.
[Bol09] Bolme D. S., Yui M. L., Draper B. A., etc. Simple real-time human detection using asingle correlation filter. Proceedings of IEEE International Workshop on PerformanceEvaluation of Tracking and Surveillance,2009,1-8.
[Cav] EC funnded CAVIAR project, IST-2001-37540,Url:http://homepages.inf.ed.ac.uk/rbf/caviar/
[Cha01] Chang C. C., Lin C. J. LIBSVM: a library for support vector machines. ACMTransactions on Intelligent Systems and Technology,2011,2(3).
[Can06] Candes E., Romberg J., Tao T. Stable signal recovery from incomplete and inaccuratemeasurements. Communications on Pure and Applied Math,2006,59(8),1207-1223.
[Col05] Collins R. T., Liu Y. X., Leordeanu M. Online selection of discriminative trackingfeatures. IEEE Transactions on Pattern Analysis and Machine Intelligence,2005,27,1631-1643.
[Com02] Comaniciu D., Meer P. Mean shift: a robust approach toward feature space analysis.IEEE Transactions on Pattern Analysis and Machine Intelligence,2002,24,603-619.
[Com03] Comaniciu D. An algorithm for data-driven bandwidth selection. IEEE Transactionson Pattern Analysis and Machine Intelligence,2003,25(2):281–288.
[Com03b] Comaniciu D., Ramesh V., Meer P. Kernel-based object tracking. IEEETransactions on Pattern Analysis and Machine Intelligence,2003,25,564-575.
[Dal05] Dalal N., Triggs B. Histograms of oriented gradients for human detection.Proceedings of IEEE Conference on Computer Vision and Pattern Recognition,2005,886-893.
[Dal06] Dalal N., Triggs B., Schmid C. Human detection using oriented histograms of fowand appearance. Proceedings of European Conference on Computer Vision,2006,2,428-441.
[Dev06] Deva Ramanan, Learning to parse images of articulated bodies. Proceedings ofConference on Advances in Neural Information Processing Systems,2006,1129-1136.
[Dol10] Dollar P., Belongie S., Perona P. The fastest pedestrian detector in the west.Proceedings of The British Machine Vision Conference,2010.
[Fan06] Fan Z., Yang M., Wu Y., etc. Efficient optimal kernel placement for reliable visualtracking. Proceedings of IEEE Conference on Computer Vision and Pattern Recognition,2006,17-22.
[Fel08] Felzenszwalb P., McAllester D., Ramanan D. A discriminatively trained, multiscale,deformable part model. Proceedings of IEEE Conference on Computer Vision and PatternRecognition,2008,1–8.
[Fer09] Ferryman J., Shahrokni A. An overview of the pets2009challenge, Proceedings ofIEEE International Workshop on Performance Evaluation of Tracking and Surveillance,2009,25-30.
[Fis81] Fischler M. A., Bolles R. C. Random sample consensus: a paradigm for model fittingwith applications to image analysis and automated cartographys. Communications of theACM,1981,24,381-395.
[Fre95] Freund Y., Schapire R. E. A decision-theoretic generalization of on-line learning andan application to boosting. Proceedings of European Conference on Computational LearningTheory,1995,23-37
[Gav98] Gavrila D. M. Multi-feature hierarchical template matching using distancetransforms. Proceedings of International Conference on Pattern Recognition,1998,1,439-444
[Gra07] Grabner M., Grabner H., Bischof H. Learning Features for Tracking. Proceedings ofIEEE Conference on Computer Vision and Pattern Recognition,2007,1-8.
[Hag98] Hager G. D., Belhumeur P. N. Efficient region tracking with parametric models ofgeometry and illumination. IEEE Transactions on Pattern Analysis and Machine Intelligence,1998,20,1025-1039.
[He09] He W., Yamashita T., Lu H. T., etc. SURF tracking. Proceedings of IEEE InternationalConference on Computer Vision,2009,1586-1592.
[Hua06] Huang C., Ai H. Z., Li Y., etc. Learning sparse features in granular space formulti-view face detection. Proceedings of International Conference on Automatic Face andGesture Recognition,2006,401-406.
[Hua07] Huang C., Ai H. Z., Li Y., etc. High-performance rotation invariant multi-view facedetection. IEEE Transactions on Pattern Analysis and Machine Intelligence,2007,29(4),671-686.
[Hua08] Huang C., Wu B., Nevatia R. Robust object tracking by hierarchical association ofdetection responses. Proceedings of European Conference on Computer Vision,2008,788-801.
[Hua10] Huang C., Nevatia R. High performance object detection by collaborative learning ofJoint Ranking of Granules features. Proceedings of IEEE Conference on Computer Vision andPattern Recognition,2010,41-48.
[Isa04] Isard M., MacCormick J. BraMBLe: a bayesian multiple-blob tracker. Proceedings ofIEEE International Conference on Computer Vision,2001,34–41.
[Isa96] Isard M., Blake A. Contour tracking by stochastic propagation of conditional density.Proceedings of European Conference on Computer Vision,1996,1,343-356.
[Isa98] Isard M., Blake A. CONDENSATION—Conditional density propagation for visualtracking. International Journal of Computer Vision,1998,29,5-28.[Jef04] Ho J., Lee K. C., Yang M. H., etc. Visual Tracking Using Learned Linear Subspaces.Proceedings of IEEE International Conference of Computer Vision and Pattern Recognition,2004,1,782-789.[Jep03] Jepson A., Fleet D., Elmaraghi T. Robust online appearance models for visualtracking. IEEE Transactions on Pattern Analysis and Machine Intelligence,2003,25(10),1296-1311.[Jey08] Jeyakar J., Venkatesh B. R., Ramakrishnan K. R. Robust object tracking withbackground-weighted local kernels, Computer Vision and Image Understanding,2008,112,296–309.[Jia07] Jiang H., Fels S., Little J. J. A linear programming approach for multiple objecttracking, Proceedings of IEEE International Conference on Computer Vision,2007,1-8.
[Kha05] Khan Z., Balch T., Dellaert F. MCMC-based particle filtering for tracking a variablenumber of interacting targets. IEEE Transactions on Pattern Analysis and MachineIntelligence,2005,27(11),1805-1819.
[Kha09] Khan Z. H., Gu I. Y. H., W. T., etc. Joint anisotropic mean shift and consensus pointfeature correspondences for object tracking in video. Proceedings of2009InternationalConference on Multimedia and Expo,2009,1270-1273.
[Kuh55] Kuhn H. W. The Hungarian method for the assignment problem. Naval ResearchLogistics Quarterly,1955,2,83–97.
[Kwo10] Kwon J., Lee K. M. Visual tracking decomposition. Proceedings of IEEEConference on Computer Vision and Pattern Recognition,2010,1269-1276.
[Lam08] Lampert C. H., Blaschko M. B., Hofmann T. Beyond sliding windows: objectlocalization by effcient subwindow search. Proceedings of IEEE Conference on ComputerVision and Pattern Recognition,2008,1-8.
[Lei07] Leibe B., Schindler K., Gool L. V. Coupled detection and trajectory estimation formulti-object tracking, Proceedings of IEEE International Conference on Computer Vision,2007,1-8.
[Lei07b] Leibe B., Seemann E., Schiele B. Robust Object Detection with InterleavedCategorization and Segmentation. International Journal of Computer Vision,2008,77(1-3),259-289.
[Lev09] Levinshtein A., Stere A., Kutulakos K. N., etc. Turbopixels: fast superpixels usinggeometric flows. IEEE Transactions on Pattern Analysis and Machine Intelligence,2009,31(12),2290-2297.
[Li04] Li S. Z., Zhang Z. Q. FloatBoost learning and statistical face detection. IEEETransactions on Pattern Analysis and Machine Intelligence,2004,26(9),1-12.
[Li07] Li Y., Ai H., Yamashita T., etc. Tracking in low frame rate video: a cascade particlefilter with discriminative observers of different lifespans, Proceedings of IEEE Conference onComputer Vision and Pattern Recognition,2007.
[Li08] Li Y., Ai H., Yamashita T., etc. Tracking in low frame rate video: a cascade particlefilter with discriminative observers of different life spans, IEEE Transactions on PatternAnalysis and Machine Intelligence,2008,30,1728-1740.
[Li12] Li W., Yang O., Wang X. G. A discriminative deep model for pedestrian detectionwith occlusion handling. Proceedings of IEEE Conference on Computer Vision and PatternRecognition,2012,3258-3265.
[Lin91] Lin J. Divergence measures based on the shannon entropy. IEEE Transactions onInformation Theory,1991,37,145-151.
[Lin04] Lin R. S., Ross D. A., Lim J., etc. Adaptive discriminative generative model and itsapplications. Proceedings of Conference on Advances in Neural Information ProcessingSystems,2004
[Liu10]刘坤.视频监控系统的关键技术研究[硕士论文].上海:上海交通大学.2010.
[Low04] Lowe D. G. Distinctive image features from scale-invariant keypoints. InternationalJournal of Computer Vision,2004,60(2),91–110.
[Loz06] Loza A., Mihaylova L., Bull D. R., etc. Structural similarity-based object tracking invideo sequences. Proceedings of International Conference on Information Fusion,2006,10-13.
[Luc81] Lucas B., Kanade T. An iterative image registration technique with an application tostereo vision. Proceedings of the7th international joint conference on Artificial intelligence,1981,2,674-679.
[Maj08] Maji S., Berg A., Malik J. Classification using intersection kernel support vectormachines is efficient. Proceedings of IEEE Conference on Computer Vision and PatternRecognition,2008,1-8.
[Mei09] Mei X., Ling H. B. Robust visual tracking using L1minimization. Proceedings ofIEEE International Conference on Computer Vision,2009,1436-1443.
[Mei11] Mei X., Ling H. B., Wu Y., etc. Minimum error bounded efficient L1tracker withocclusion detection, Proceedings of IEEE Conference on Computer Vision and PatternRecognition,2011,1257-1264.
[Mu08] Mu Y., Yan S., Liu Y., etc. Discriminative local binary patterns for human detection inpersonal album. Proceedings of IEEE Conference on Computer Vision and PatternRecognition,2008,1–8.
[Ngu01] Nguyen H.T., Worring M., Van Den Boomgaard R. Occlusion robust adaptivetemplate tracking. Proceedings of IEEE International Conference on Computer Vision,2001,1,678–683.
[Oja98] Ojala T., Pietikinen M., Harwood D. A comparative study of texture measures withclassification based on feature distributions. Pattern Recognition,1998,29(1),51–59.
[Oja02] Ojala T., Pietikainen M., Maenpaa T. Multiresolution gray-scale and rotationinvariant texture classification with local binary patterns. IEEE Transactions on PatternAnalysis and Machine Intelligence,2002,24(7),971-987.
[Oku04] Okuma K., Taleghani A., Freitas N. D., etc. A boosted particle filter: multitargetdetection and tracking, Proceedings of European Conference on Computer Vision,1998,28-39.
[Ope] Open Source Computer Vision, Url: http://opencv.willowgarage.com/wiki/
[Oro12] Oron, S., Bar-Hillel A., Levi D., etc. Locally Orderless Tracking. Proceedings ofIEEE Conference on Computer Vision and Pattern Recognition,2012,1,1940-1947.
[Pet07] Tenth IEEE International Workshop on Performance Evaluation of Tracking andSurveillance, datasets available at: http://pets2007.net/.
[Ros08] Ross D. A., Lim J. W., Lin R. S., etc. Incremental Learning for Robust VisualTracking. International Journal of Computer Vision.2008,77(1-3),125-141.
[Rot04] Rother G., Kolmogorov V., Blake A."GrabCut"-Interactive foreground extractionusing iterated graph cuts. ACM Transactions on Graphics,2004,23(4),309-314.
[Rot09] Roth P.M., Sternig S., Grabner H., etc. Classifier grids for robust adaptive objectdetection. Proceedings of IEEE Conference on Computer Vision and Pattern Recognition,2009,2727-2734.
[Shu12] Shu G., Dehghan A., Oreifej O., etc. Part-based Multiple-Person Tracking withPartial Occlusion Handling. Proceedings of IEEE Conference on Computer Vision and PatternRecognition,2012,1815-1821.
[Sid00] Sidenbladh H., Black M. J., Fleet D. J. Stochastic tracking of3D human figures using2D image motion. Proceedings of European conference on computer vision,2000,702-718.
[Smi05] Smith K., Perez D. G., Odobez J. M. Using particles to track varying numbers ofinteracting people. Proceedings of IEEE Conference on Computer Vision and PatternRecognition,2005,962-969.
[Sta00] Stauffer C., Grimson W. E. Learning patterns of activity using real-time tracking,IEEE Transactions on Pattern Analysis and Machine Intelligence,2000,22(8),747-757.
[Sta09] Stalder S., Grabner H., Van Gool L. Beyond semi-supervised tracking: Trackingshould be as simple as detection, but not simpler than recognition. IEEE InternationalWorkshop on Computer Vision,2009,1409-1416.
[Sum08] Sumin Q., Wu X. H. Hand tracking and gesture gecogniton by anisotropic kernelmean shift, Proceedings of International Conference on Neural Networks and SignalProcessing,2008,581-585.
[Tom91] Tomasi C., Kanade T. Detection and tracking of point features. Technical ReportCMU-CS-91-132, Carnegie Mellon University,1991.
[Tuz07] Tuzel O., Porikli F., Meer P. Human detection via classification on riemannianmanifold. Proceedings of IEEE Conference on Computer Vision and Pattern Recognition,2007,1-8.
[Vio01] Viola P., Jones M. Rapid object detection using a boosted cascade of simple features.Proceedings of IEEE Conference on Computer Vision and Pattern Recognition,2001,1,511-518.
[Vio03] Viola P., Jones M. J., Snow D. Detecting pedestrians using patterns of motion andappearance. Proceedings of IEEE International Conference on Computer Vision,2003,2,734–741.
[Wal10] Walk S., Majer N., Schindler K., etc. New features and insights for pedestriandetection, Proceedings of IEEE Conference on Computer Vision and Pattern Recognition,2010,1030-1037.
[Wan09a] Wang F. L., Yu S. Y., Yang J. Robust and efficient fragments-based tracking usingmean shift. International Journal of Electronics and Communications,64(7),614-623.
[Wan09b] Wang X., Han T. X., Yan S. An HOG-LBP human detector with partial occlusionhandling, Proceedings of IEEE Conference on Computer Vision and Pattern Recognition,2009,32-39.
[Wan11] Wang S., Lu H. C., Yang F. etc. Superpixel tracking. Proceedings of IEEEInternational Conference on Computer Vision,2011,1323-1330.
[Wan12] Wang M., Li W., Wang X. G. Transferring a generic pedestrian detector towardsspecific scenes. Proceedings of IEEE Conference on Computer Vision and PatternRecognition,2012,3274-3281.
[Wer09] Werlberger M., Trobin W., Pock T., etc. Anisotropic Huber-L1optical flow.Proceedings of the British Machine Vision Conference,2009.
[Wjq08] Wang J., Yagi Y. Integrating color and shape-texture features for adaptive real-timeobject tracking. IEEE Transactions on Image Processing,2007,17,235-240.
[Wu05] Wu B., Nevatia R. Detection of multiple, partially occluded humans in a single imageby bayesian combination of edgelet part detectors. Proceedings of IEEE InternationalConference on Computer Vision,2005,90-97.
[Wu06] Wu B., Nevatia R. Tracking of multiple, partially occluded humans based on staticbody part. Proceedings of IEEE Conference on Computer Vision and Pattern Recognition,2006,951-958.
[Wu07] Wu B., Nevatia R. Cluster boosted tree classifier for multi-view, multi-pose objectdetection. Proceedings of IEEE International Conference on Computer Vision,2007,1-8.
[Wu08] Wu B., Nevatia R. Optimizing discrimination-efficiency tradeoff in integratingheterogeneous local features for object detection. Proceedings of IEEE Conference onComputer Vision and Pattern Recognition,2008,1–8.
[Wu11]吴亚丽，张俊敏.运筹学.北京:北京大学出版社.2011:79-82.
[Yan09] Yang M., Lv F., Xu W., etc. Detection driven adaptive multi-cue integration formultiple human tracking, Proceedings of IEEE International Conference on Computer Vision,2009,1554-1561.
[Yil03] Yilmaz A., Shafique K., Shah M. Target tracking in airborne forward looking infraredimagery. Image and Vision Computing.2003,21,623-635.
[Zha04] Zhao T., Nevatia R. Tracking multiple humans in complex situations. IEEETransactions on Pattern Analysis and Machine Intelligence,2004,26(9),1208-1221.
[Zha08] Zhang L., Li Y., Nevatia R. Global data association for multi-object tracking usingnetwork flows. Proceedings of IEEE International Conference on Computer Vision,2008,1-8.
[Zha12] Zhang K. H., Zhang L., Yang M. H. Real-Time Compressive Tracking.Proceedings ofEuropean Conference on Computer Vision,2012,864-877
[Zhe07] Zhe L., Davis L. S., Doermann D., etc. Hierarchical part-template matching forhuman detection and segmentation. Proceedings of IEEE International Conference onComputer Vision,2007,1-8.
[Zhu06] Zhu Q., Yeh M. C., Cheng K.T., etc. Fast human detection using a cascade of
histograms of oriented gradients. Proceedings of IEEE Conference on Computer Vision and
Pattern Recognition,2006,1491-1498.
[Aga02] Agarwal S., Roth D. Learning a sparse representation for object detection.Proceedings of European Conference on Computer Vision,2002,4,113-127.
[And09] Andriluka M., Roth S., Schiele B. Pictorial structures revisited: People detection andarticulated pose estimation. Proceedings of IEEE Conference on Computer Vision and PatternRecognition,2009,1014-1021.
[Bab09] Babenko B., Yang M. H., Belongie S. Visual tracking with online Multiple InstanceLearning. Proceedings of IEEE Conference on Computer Vision and Pattern Recognition,2009,983-990.
[Bay06] Bay H., Tuytelaars T., Gool L. J. Surf: Speeded up robust features. Proceedings ofEuropean Conference on Computer Vision,2006,1,404-417.
[Bel09] Beleznai C., Bischof H. Fast human detection in crowded scenes by contourintegration and local shape estimation. Proceedings of IEEE Conference on Computer Visionand Pattern Recognition,2009,2246-2253.
[Ben11a] Benfold B., Reid I. D. Stable Multi-Target Tracking in Real-Time SurveillanceVideo. Proceedings of IEEE Conference on Computer Vision and Pattern Recognition,2011,3457-3464.
[Ben11b] Benenson R., Timofte R., Gool v. L. Stixels estimation without depthmapcomputation. Proceedings of the Computer Vision in Vehicle Technology Workshop,2011.
[Ben12] Benenson R., Mathias M., Timofte R., etc. Pedestrian detection at100frames persecond.Proceedings of IEEE Conference on Computer Vision and Pattern Recognition,2012,2903-2910.
[Ber06] Berclaz J., Fleuret F., Fua P. Robust people tracking with global trajectoryoptimization. Proceedings of IEEE International Conference on Computer Vision,2006,1,744-750.
[Ber08] Bernardin K., Stiefelhagen R. Evaluating multipe object tracking performance: theCLEARMOT metrics. EURASIP Journal on Image and Video Processing,2008.
[Bla98] Black M. J., Jepson A. D. EigenTracking: Robust matching and tracking ofarticulated objects using a view-based representation. International Journal of ComputerVision,1998,26,63-84.
[Bre09] Breitenstein M. D., Reichlin F., Leibe B., etc. Robust tracking-by-detection using adetector confidence particle filter. Proceedings of IEEE International Conference onComputer Vision,2009,1515-1522.
[Bol09] Bolme D. S., Yui M. L., Draper B. A., etc. Simple real-time human detection using asingle correlation filter. Proceedings of IEEE International Workshop on PerformanceEvaluation of Tracking and Surveillance,2009,1-8.
[Cav] EC funnded CAVIAR project, IST-2001-37540,Url:http://homepages.inf.ed.ac.uk/rbf/caviar/
[Cha01] Chang C. C., Lin C. J. LIBSVM: a library for support vector machines. ACMTransactions on Intelligent Systems and Technology,2011,2(3).
[Can06] Candes E., Romberg J., Tao T. Stable signal recovery from incomplete and inaccuratemeasurements. Communications on Pure and Applied Math,2006,59(8),1207-1223.
[Col05] Collins R. T., Liu Y. X., Leordeanu M. Online selection of discriminative trackingfeatures. IEEE Transactions on Pattern Analysis and Machine Intelligence,2005,27,1631-1643.
[Com02] Comaniciu D., Meer P. Mean shift: a robust approach toward feature space analysis.IEEE Transactions on Pattern Analysis and Machine Intelligence,2002,24,603-619.
[Com03] Comaniciu D. An algorithm for data-driven bandwidth selection. IEEE Transactionson Pattern Analysis and Machine Intelligence,2003,25(2):281–288.
[Com03b] Comaniciu D., Ramesh V., Meer P. Kernel-based object tracking. IEEETransactions on Pattern Analysis and Machine Intelligence,2003,25,564-575.
[Dal05] Dalal N., Triggs B. Histograms of oriented gradients for human detection.Proceedings of IEEE Conference on Computer Vision and Pattern Recognition,2005,886-893.
[Dal06] Dalal N., Triggs B., Schmid C. Human detection using oriented histograms of fowand appearance. Proceedings of European Conference on Computer Vision,2006,2,428-441.
[Dev06] Deva Ramanan, Learning to parse images of articulated bodies. Proceedings ofConference on Advances in Neural Information Processing Systems,2006,1129-1136.
[Dol10] Dollar P., Belongie S., Perona P. The fastest pedestrian detector in the west.Proceedings of The British Machine Vision Conference,2010.
[Fan06] Fan Z., Yang M., Wu Y., etc. Efficient optimal kernel placement for reliable visualtracking. Proceedings of IEEE Conference on Computer Vision and Pattern Recognition,2006,17-22.
[Fel08] Felzenszwalb P., McAllester D., Ramanan D. A discriminatively trained, multiscale,deformable part model. Proceedings of IEEE Conference on Computer Vision and PatternRecognition,2008,1–8.
[Fer09] Ferryman J., Shahrokni A. An overview of the pets2009challenge, Proceedings ofIEEE International Workshop on Performance Evaluation of Tracking and Surveillance,2009,25-30.
[Fis81] Fischler M. A., Bolles R. C. Random sample consensus: a paradigm for model fittingwith applications to image analysis and automated cartographys. Communications of theACM,1981,24,381-395.
[Fre95] Freund Y., Schapire R. E. A decision-theoretic generalization of on-line learning andan application to boosting. Proceedings of European Conference on Computational LearningTheory,1995,23-37
[Gav98] Gavrila D. M. Multi-feature hierarchical template matching using distancetransforms. Proceedings of International Conference on Pattern Recognition,1998,1,439-444
[Gra07] Grabner M., Grabner H., Bischof H. Learning Features for Tracking. Proceedings ofIEEE Conference on Computer Vision and Pattern Recognition,2007,1-8.
[Hag98] Hager G. D., Belhumeur P. N. Efficient region tracking with parametric models ofgeometry and illumination. IEEE Transactions on Pattern Analysis and Machine Intelligence,1998,20,1025-1039.
[He09] He W., Yamashita T., Lu H. T., etc. SURF tracking. Proceedings of IEEE InternationalConference on Computer Vision,2009,1586-1592.
[Hua06] Huang C., Ai H. Z., Li Y., etc. Learning sparse features in granular space formulti-view face detection. Proceedings of International Conference on Automatic Face andGesture Recognition,2006,401-406.
[Hua07] Huang C., Ai H. Z., Li Y., etc. High-performance rotation invariant multi-view facedetection. IEEE Transactions on Pattern Analysis and Machine Intelligence,2007,29(4),671-686.
[Hua08] Huang C., Wu B., Nevatia R. Robust object tracking by hierarchical association ofdetection responses. Proceedings of European Conference on Computer Vision,2008,788-801.
[Hua10] Huang C., Nevatia R. High performance object detection by collaborative learning ofJoint Ranking of Granules features. Proceedings of IEEE Conference on Computer Vision andPattern Recognition,2010,41-48.
[Isa04] Isard M., MacCormick J. BraMBLe: a bayesian multiple-blob tracker. Proceedings ofIEEE International Conference on Computer Vision,2001,34–41.
[Isa96] Isard M., Blake A. Contour tracking by stochastic propagation of conditional density.Proceedings of European Conference on Computer Vision,1996,1,343-356.
[Isa98] Isard M., Blake A. CONDENSATION—Conditional density propagation for visualtracking. International Journal of Computer Vision,1998,29,5-28.[Jef04] Ho J., Lee K. C., Yang M. H., etc. Visual Tracking Using Learned Linear Subspaces.Proceedings of IEEE International Conference of Computer Vision and Pattern Recognition,2004,1,782-789.[Jep03] Jepson A., Fleet D., Elmaraghi T. Robust online appearance models for visualtracking. IEEE Transactions on Pattern Analysis and Machine Intelligence,2003,25(10),1296-1311.[Jey08] Jeyakar J., Venkatesh B. R., Ramakrishnan K. R. Robust object tracking withbackground-weighted local kernels, Computer Vision and Image Understanding,2008,112,296–309.[Jia07] Jiang H., Fels S., Little J. J. A linear programming approach for multiple objecttracking, Proceedings of IEEE International Conference on Computer Vision,2007,1-8.
[Kha05] Khan Z., Balch T., Dellaert F. MCMC-based particle filtering for tracking a variablenumber of interacting targets. IEEE Transactions on Pattern Analysis and MachineIntelligence,2005,27(11),1805-1819.
[Kha09] Khan Z. H., Gu I. Y. H., W. T., etc. Joint anisotropic mean shift and consensus pointfeature correspondences for object tracking in video. Proceedings of2009InternationalConference on Multimedia and Expo,2009,1270-1273.
[Kuh55] Kuhn H. W. The Hungarian method for the assignment problem. Naval ResearchLogistics Quarterly,1955,2,83–97.
[Kwo10] Kwon J., Lee K. M. Visual tracking decomposition. Proceedings of IEEEConference on Computer Vision and Pattern Recognition,2010,1269-1276.
[Lam08] Lampert C. H., Blaschko M. B., Hofmann T. Beyond sliding windows: objectlocalization by effcient subwindow search. Proceedings of IEEE Conference on ComputerVision and Pattern Recognition,2008,1-8.
[Lei07] Leibe B., Schindler K., Gool L. V. Coupled detection and trajectory estimation formulti-object tracking, Proceedings of IEEE International Conference on Computer Vision,2007,1-8.
[Lei07b] Leibe B., Seemann E., Schiele B. Robust Object Detection with InterleavedCategorization and Segmentation. International Journal of Computer Vision,2008,77(1-3),259-289.
[Lev09] Levinshtein A., Stere A., Kutulakos K. N., etc. Turbopixels: fast superpixels usinggeometric flows. IEEE Transactions on Pattern Analysis and Machine Intelligence,2009,31(12),2290-2297.
[Li04] Li S. Z., Zhang Z. Q. FloatBoost learning and statistical face detection. IEEETransactions on Pattern Analysis and Machine Intelligence,2004,26(9),1-12.
[Li07] Li Y., Ai H., Yamashita T., etc. Tracking in low frame rate video: a cascade particlefilter with discriminative observers of different lifespans, Proceedings of IEEE Conference onComputer Vision and Pattern Recognition,2007.
[Li08] Li Y., Ai H., Yamashita T., etc. Tracking in low frame rate video: a cascade particlefilter with discriminative observers of different life spans, IEEE Transactions on PatternAnalysis and Machine Intelligence,2008,30,1728-1740.
[Li12] Li W., Yang O., Wang X. G. A discriminative deep model for pedestrian detectionwith occlusion handling. Proceedings of IEEE Conference on Computer Vision and PatternRecognition,2012,3258-3265.
[Lin91] Lin J. Divergence measures based on the shannon entropy. IEEE Transactions onInformation Theory,1991,37,145-151.
[Lin04] Lin R. S., Ross D. A., Lim J., etc. Adaptive discriminative generative model and itsapplications. Proceedings of Conference on Advances in Neural Information ProcessingSystems,2004
[Liu10]刘坤.视频监控系统的关键技术研究[硕士论文].上海:上海交通大学.2010.
[Low04] Lowe D. G. Distinctive image features from scale-invariant keypoints. InternationalJournal of Computer Vision,2004,60(2),91–110.
[Loz06] Loza A., Mihaylova L., Bull D. R., etc. Structural similarity-based object tracking invideo sequences. Proceedings of International Conference on Information Fusion,2006,10-13.
[Luc81] Lucas B., Kanade T. An iterative image registration technique with an application tostereo vision. Proceedings of the7th international joint conference on Artificial intelligence,1981,2,674-679.
[Maj08] Maji S., Berg A., Malik J. Classification using intersection kernel support vectormachines is efficient. Proceedings of IEEE Conference on Computer Vision and PatternRecognition,2008,1-8.
[Mei09] Mei X., Ling H. B. Robust visual tracking using L1minimization. Proceedings ofIEEE International Conference on Computer Vision,2009,1436-1443.
[Mei11] Mei X., Ling H. B., Wu Y., etc. Minimum error bounded efficient L1tracker withocclusion detection, Proceedings of IEEE Conference on Computer Vision and PatternRecognition,2011,1257-1264.
[Mu08] Mu Y., Yan S., Liu Y., etc. Discriminative local binary patterns for human detection inpersonal album. Proceedings of IEEE Conference on Computer Vision and PatternRecognition,2008,1–8.
[Ngu01] Nguyen H.T., Worring M., Van Den Boomgaard R. Occlusion robust adaptivetemplate tracking. Proceedings of IEEE International Conference on Computer Vision,2001,1,678–683.
[Oja98] Ojala T., Pietikinen M., Harwood D. A comparative study of texture measures withclassification based on feature distributions. Pattern Recognition,1998,29(1),51–59.
[Oja02] Ojala T., Pietikainen M., Maenpaa T. Multiresolution gray-scale and rotationinvariant texture classification with local binary patterns. IEEE Transactions on PatternAnalysis and Machine Intelligence,2002,24(7),971-987.
[Oku04] Okuma K., Taleghani A., Freitas N. D., etc. A boosted particle filter: multitargetdetection and tracking, Proceedings of European Conference on Computer Vision,1998,28-39.
[Ope] Open Source Computer Vision, Url: http://opencv.willowgarage.com/wiki/
[Oro12] Oron, S., Bar-Hillel A., Levi D., etc. Locally Orderless Tracking. Proceedings ofIEEE Conference on Computer Vision and Pattern Recognition,2012,1,1940-1947.
[Pet07] Tenth IEEE International Workshop on Performance Evaluation of Tracking andSurveillance, datasets available at: http://pets2007.net/.
[Ros08] Ross D. A., Lim J. W., Lin R. S., etc. Incremental Learning for Robust VisualTracking. International Journal of Computer Vision.2008,77(1-3),125-141.
[Rot04] Rother G., Kolmogorov V., Blake A."GrabCut"-Interactive foreground extractionusing iterated graph cuts. ACM Transactions on Graphics,2004,23(4),309-314.
[Rot09] Roth P.M., Sternig S., Grabner H., etc. Classifier grids for robust adaptive objectdetection. Proceedings of IEEE Conference on Computer Vision and Pattern Recognition,2009,2727-2734.
[Shu12] Shu G., Dehghan A., Oreifej O., etc. Part-based Multiple-Person Tracking withPartial Occlusion Handling. Proceedings of IEEE Conference on Computer Vision and PatternRecognition,2012,1815-1821.
[Sid00] Sidenbladh H., Black M. J., Fleet D. J. Stochastic tracking of3D human figures using2D image motion. Proceedings of European conference on computer vision,2000,702-718.
[Smi05] Smith K., Perez D. G., Odobez J. M. Using particles to track varying numbers ofinteracting people. Proceedings of IEEE Conference on Computer Vision and PatternRecognition,2005,962-969.
[Sta00] Stauffer C., Grimson W. E. Learning patterns of activity using real-time tracking,IEEE Transactions on Pattern Analysis and Machine Intelligence,2000,22(8),747-757.
[Sta09] Stalder S., Grabner H., Van Gool L. Beyond semi-supervised tracking: Trackingshould be as simple as detection, but not simpler than recognition. IEEE InternationalWorkshop on Computer Vision,2009,1409-1416.
[Sum08] Sumin Q., Wu X. H. Hand tracking and gesture gecogniton by anisotropic kernelmean shift, Proceedings of International Conference on Neural Networks and SignalProcessing,2008,581-585.
[Tom91] Tomasi C., Kanade T. Detection and tracking of point features. Technical ReportCMU-CS-91-132, Carnegie Mellon University,1991.
[Tuz07] Tuzel O., Porikli F., Meer P. Human detection via classification on riemannianmanifold. Proceedings of IEEE Conference on Computer Vision and Pattern Recognition,2007,1-8.
[Vio01] Viola P., Jones M. Rapid object detection using a boosted cascade of simple features.Proceedings of IEEE Conference on Computer Vision and Pattern Recognition,2001,1,511-518.
[Vio03] Viola P., Jones M. J., Snow D. Detecting pedestrians using patterns of motion andappearance. Proceedings of IEEE International Conference on Computer Vision,2003,2,734–741.
[Wal10] Walk S., Majer N., Schindler K., etc. New features and insights for pedestriandetection, Proceedings of IEEE Conference on Computer Vision and Pattern Recognition,2010,1030-1037.
[Wan09a] Wang F. L., Yu S. Y., Yang J. Robust and efficient fragments-based tracking usingmean shift. International Journal of Electronics and Communications,64(7),614-623.
[Wan09b] Wang X., Han T. X., Yan S. An HOG-LBP human detector with partial occlusionhandling, Proceedings of IEEE Conference on Computer Vision and Pattern Recognition,2009,32-39.
[Wan11] Wang S., Lu H. C., Yang F. etc. Superpixel tracking. Proceedings of IEEEInternational Conference on Computer Vision,2011,1323-1330.
[Wan12] Wang M., Li W., Wang X. G. Transferring a generic pedestrian detector towardsspecific scenes. Proceedings of IEEE Conference on Computer Vision and PatternRecognition,2012,3274-3281.
[Wer09] Werlberger M., Trobin W., Pock T., etc. Anisotropic Huber-L1optical flow.Proceedings of the British Machine Vision Conference,2009.
[Wjq08] Wang J., Yagi Y. Integrating color and shape-texture features for adaptive real-timeobject tracking. IEEE Transactions on Image Processing,2007,17,235-240.
[Wu05] Wu B., Nevatia R. Detection of multiple, partially occluded humans in a single imageby bayesian combination of edgelet part detectors. Proceedings of IEEE InternationalConference on Computer Vision,2005,90-97.
[Wu06] Wu B., Nevatia R. Tracking of multiple, partially occluded humans based on staticbody part. Proceedings of IEEE Conference on Computer Vision and Pattern Recognition,2006,951-958.
[Wu07] Wu B., Nevatia R. Cluster boosted tree classifier for multi-view, multi-pose objectdetection. Proceedings of IEEE International Conference on Computer Vision,2007,1-8.
[Wu08] Wu B., Nevatia R. Optimizing discrimination-efficiency tradeoff in integratingheterogeneous local features for object detection. Proceedings of IEEE Conference onComputer Vision and Pattern Recognition,2008,1–8.
[Wu11]吴亚丽，张俊敏.运筹学.北京:北京大学出版社.2011:79-82.
[Yan09] Yang M., Lv F., Xu W., etc. Detection driven adaptive multi-cue integration formultiple human tracking, Proceedings of IEEE International Conference on Computer Vision,2009,1554-1561.
[Yil03] Yilmaz A., Shafique K., Shah M. Target tracking in airborne forward looking infraredimagery. Image and Vision Computing.2003,21,623-635.
[Zha04] Zhao T., Nevatia R. Tracking multiple humans in complex situations. IEEETransactions on Pattern Analysis and Machine Intelligence,2004,26(9),1208-1221.
[Zha08] Zhang L., Li Y., Nevatia R. Global data association for multi-object tracking usingnetwork flows. Proceedings of IEEE International Conference on Computer Vision,2008,1-8.
[Zha12] Zhang K. H., Zhang L., Yang M. H. Real-Time Compressive Tracking.Proceedings ofEuropean Conference on Computer Vision,2012,864-877
[Zhe07] Zhe L., Davis L. S., Doermann D., etc. Hierarchical part-template matching forhuman detection and segmentation. Proceedings of IEEE International Conference onComputer Vision,2007,1-8.
[Zhu06] Zhu Q., Yeh M. C., Cheng K.T., etc. Fast human detection using a cascade of
histograms of oriented gradients. Proceedings of IEEE Conference on Computer Vision and
Pattern Recognition,2006,1491-1498.