复杂场景下视觉目标跟踪方法研究
|
摘要
视觉目标跟踪是计算机视觉领域最重要、最热门的研究问题之一。它融合了图像处理、模式识别、人工智能、图像内容理解等相关领域的多种知识。它是一种从图像序列中检测目标,获得目标状态,自动跟踪目标的技术。视觉跟踪的应用领域非常广泛,可用于目标检测、目标识别、图像内容理解、场景分类、行为理解,人机互动、机器人导航等领域。复杂场景下的单目标跟踪和多目标跟踪是视觉目标跟踪中最困难的问题一直没有得到很好的解决。
复杂场景下目标跟踪由于其跟踪场景的复杂性导致它有更高的难度。在复杂场景下由于在光照强度变化,背景的干扰和同一个场景下各个跟踪对象之间具有强相似性等特点,所以目标特征描述很困难。另一个方面由于复杂场景下目标运动状态的非线性、非高斯、多模态、高噪声等特点对跟踪算法也有了更严格的要求。基于粒子滤波的跟踪方法成为复杂场景目标跟踪的主流方法。但是粒子滤波也有粒子数目越多,算法复杂度越高,由于重采样使具有较大权重的样本被多次选取,导致采样结果中样本多样性下降,使得粒子滤波出现粒子贫乏问题。在多目标跟踪中的一个难点就是目标关联概率的估计。
针对上述问题本文的工作内容有:在复杂场景下本文利用颜色和梯度方向双重信息描述目标。在多目标跟踪中采用SIFT特征描述。将粒子群优化算法(PSO)引入粒子滤波对粒子进行优化,使粒子朝着后验概率密度大的区域移动从而减少所需样本,同时增加一个多样化步骤增加样本的多样性。提出一种新的PSO-PF图像跟踪算法。在多目标跟踪中提出一种基于条件随机场的关联概率估计方法,实现复杂场景下的多目标跟踪。
Visual target tracking is one of the most important and popular field of computer vision. It combines image processing, pattern recognition, artificial intelligence, image content understanding and other related fields of knowledge. It is a kind of technology that detect target from image sequences, obtain target state, automatic target tracking. It can be used for target detection, target recognition, image content understanding, scene classification, behavior understanding, human-computer interaction, robot navigation and other fields. Complex scenarios single target tracking and multi-target tracking is the most difficult visual target tracking problem has not been solved.
Target tracking under complex scene due to the complexity of tracking scene result it has a higher degree of difficulty. In the complex scenario because changes in light intensity, background interference and the same scenario objects have strong similarity, so target characterization is very difficult. As the target motion state in complex scenes with non-linear, non-Gaussian, multi-mode, high noise, and so tracking algorithm also has more stringent requirements. Tracking method based on particle filter is the main tracking method in complex scenes. However, the number of particles in particle filter is also the more higher the complexity of the algorithm, since resampling to a large sample weights are selected many times, resulting in decreased diversity of sampling results, making particle filter poor problem occurs. In the multi-target tracking is a difficult target association probability estimates.
Address these issues my work includes: Color and gradient direction dual information were used to describe in complex scenarios. SIFT feature was used to describe object in multi-target tracking. The particle swarm optimization (PSO) was introduced to optimize the particle filter. The particles move toward the region of larger posterior probability density to reduce the required samples, while increasing the diversity of samples. This paper presents a new image tracking algorithm PSO-PF.A new association probability estimation method was proposed based on conditional random field in the multi-target tracking.
复杂场景下目标跟踪由于其跟踪场景的复杂性导致它有更高的难度。在复杂场景下由于在光照强度变化,背景的干扰和同一个场景下各个跟踪对象之间具有强相似性等特点,所以目标特征描述很困难。另一个方面由于复杂场景下目标运动状态的非线性、非高斯、多模态、高噪声等特点对跟踪算法也有了更严格的要求。基于粒子滤波的跟踪方法成为复杂场景目标跟踪的主流方法。但是粒子滤波也有粒子数目越多,算法复杂度越高,由于重采样使具有较大权重的样本被多次选取,导致采样结果中样本多样性下降,使得粒子滤波出现粒子贫乏问题。在多目标跟踪中的一个难点就是目标关联概率的估计。
针对上述问题本文的工作内容有:在复杂场景下本文利用颜色和梯度方向双重信息描述目标。在多目标跟踪中采用SIFT特征描述。将粒子群优化算法(PSO)引入粒子滤波对粒子进行优化,使粒子朝着后验概率密度大的区域移动从而减少所需样本,同时增加一个多样化步骤增加样本的多样性。提出一种新的PSO-PF图像跟踪算法。在多目标跟踪中提出一种基于条件随机场的关联概率估计方法,实现复杂场景下的多目标跟踪。
Visual target tracking is one of the most important and popular field of computer vision. It combines image processing, pattern recognition, artificial intelligence, image content understanding and other related fields of knowledge. It is a kind of technology that detect target from image sequences, obtain target state, automatic target tracking. It can be used for target detection, target recognition, image content understanding, scene classification, behavior understanding, human-computer interaction, robot navigation and other fields. Complex scenarios single target tracking and multi-target tracking is the most difficult visual target tracking problem has not been solved.
Target tracking under complex scene due to the complexity of tracking scene result it has a higher degree of difficulty. In the complex scenario because changes in light intensity, background interference and the same scenario objects have strong similarity, so target characterization is very difficult. As the target motion state in complex scenes with non-linear, non-Gaussian, multi-mode, high noise, and so tracking algorithm also has more stringent requirements. Tracking method based on particle filter is the main tracking method in complex scenes. However, the number of particles in particle filter is also the more higher the complexity of the algorithm, since resampling to a large sample weights are selected many times, resulting in decreased diversity of sampling results, making particle filter poor problem occurs. In the multi-target tracking is a difficult target association probability estimates.
Address these issues my work includes: Color and gradient direction dual information were used to describe in complex scenarios. SIFT feature was used to describe object in multi-target tracking. The particle swarm optimization (PSO) was introduced to optimize the particle filter. The particles move toward the region of larger posterior probability density to reduce the required samples, while increasing the diversity of samples. This paper presents a new image tracking algorithm PSO-PF.A new association probability estimation method was proposed based on conditional random field in the multi-target tracking.
引文
[1]候志强,韩崇昭.视觉跟踪技术综述.自动化学报2006, 32(4): 603-618
[2] CollinsR.T.,etal . A system for Video Surveillance and Monitoring, CMU-RI-TR
[3] http://www-sop.inria.fr/orion/ADVISOR
[4] Matsuyama T. Cooperative Distributed Vision. In Proceedings of DARPA Image Understanding Workshop, 1998:365-384
[5] D.Comaniciu, V.Ramesh and P.Meer. Kernel-based object tracking. IEEE Transaction on Pattern Analysis and Machine Intelligence, 2003, 25(5): 564-577
[6] C.Yang, R.Duraiswami and L.Davis. Efficient mean-shift tracking via a new similarity measure. In:Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition, San Diego, CA, United States, 2005, I, 176-183
[7]彭宁嵩,杨杰,刘志等. Mean-Shift跟踪算法中核函数窗宽的自动选取.软件学报, 2005,16(9): 1542-1551
[8] J.Tu, H.Tao and T.Huang.Online updating appearance generative mixture model for meanshift tracking. In: Proceedings of Asian Conference on Computer Vision, Hyderabad,India, 2006, 1, 694-703
[9] Y.Chen,Y.Rui and T.S.Huang. JPDAF based HMM for real-time contour tracking. In: Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Kauai, Hawaii, 2001, 1, 543-550
[10] Xinyu Xu and Baoxin Li. Rao-Blackwellised Particle Filter for Tracking with Application in Visual Surveillance. Proceedings Joint IEEE International Workshop on VS-PETS, Beijing, October 15-18, 2005
[11] Swain, M.J, Ballard, D.H. Color indexing. International Journal of Computer Vision, 1991, 7(1):11~32
[12] Stricker, M., Orengo, M. Similarity of color images. In: Niblack, W.R., Rceds, J., eds. Proceedings of the SPIE2420, Storage and Retrieval for Image and Video DatabaseIII. San Jose, CA: SPIE, 1995. 381~392
[13] Liu, Zhong-wei. Color image retrieval using local accumulative histogram. Journal ofimage and graphics, 1998, 3(7): 533~537
[14] D. Androutsos, K.N.Plataniotis, A.N.Venetsanopoulos. A Novel vector-based approach to color image retrieval using a vector angular-based distance measure. Computer Vision and Image Understanding, 1999, 75(1/2): 46~58
[15] Pass G, Zabih R. Histogram refinement for content-based image retrieval. In: Proceedings of the 3rd IEEE Workshop on Applications of Computer Vision (WACV’96). Sarasota: IEEE Computer Society, 1996. 96~102
[18] Huang J. Image indexing using color correlograms. In: Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. San Juan: IEEE Computer Society, 1997. 762~768
[19] D.G.Lowe. Distinctive image features from scale-invariant keyPoints. IJCV, 60(2): 91-110 , 2004
[20] NUmmiaroK, Koller-Meier, Van GOOLL. Color features for tracking non-rigidobjeets,Special Issue on Visual Surveillanee, Chinese Journal of Automation,2003: 345-355
[21] Arulampalam M S, Marskell S, Gordon N, et al. A tutorial on particle filters for online nonlinear/non-Gaussian Bayesian tracking.IEEE Transactions On Signal Processing, 2002, 50(2): 174-188
[22] McGinnity S, Irwin G. Multiple model estimation using the bootstrap filter. IEEE Colloquium on Target Tracking and Data Fusion , 1998: 3/1-3/3
[23] Liu J.S and Chen R. Sequential Monte Carlo Methods for Dynamical Systems. Journal of the American Statistical Association, 1998, 93(5): 1032-1044
[24] Kitagawa G. Monte Carlo Filter and Smoother for Non-Gaussian Nonlinear State Space Models. Journal of Computational and Graphical Statistics, 1996, 5(1): 1-25
[25] Kennedy J, Eberhart R. Particle swarm optimization. Proceedings of the IEEE International Conference on Neural Network Piscataway, NJ, USA: IEEE, 1995: 1942– 1948
[26] R.Van Merwe, A.Doucet, N.de Freitas, et al. The unscented particle filter, Technical report CUED/F-INFENG/TR380, Cambridge University Engineering Department, August, 2000
[27] N.J.Gordon, D.J.Salmond and A.F.M.Smith. Novel approach to nonlinear/non-GaussianBayesian state estimation. IEEE Proceedings-F of Radar and Signal Processing, 1993, 140(2): 107-113
[28] T.Higuchi. Monte Carlo filter using the genetic algorithm operators.Journal of Statistical Computation and Simulation, 1997, 59(1): 1-23
[29] J.S.Liu and R.Chen. Sequential Monte Carlo methods for dynamic systems. Journal of the American Statistical Association, 1998, 93(443): 1032-1044
[30] G.Kitagawa. Monte carlo filter and smoother for non-gaussian nonlinear state space models. Journal of Computational and Graphical Statistics, 1996, 5: 1-25
[31] J.Kennedy and R.C.Eberhart. Particle swarm optimization , in Proc. IEEE Int’l Conf.on Neural Networks, pp.1942-1948, 1995
[32] M.Clerc and J.kennedy. The particle swarm-explosion, stability and convergence in a multidimensional complex space. IEEE Trans.on Evolutionary Computation , 6(1): 58-73, 2002
[33] M.P.Wachowiak, R.Smolikova and Y.Zheng. An approach to multimodal biomedical image registration utilizing particle swarm optimization. IEEE Trans.on Evolutionary Computation, 8(3): 289-301, 2004
[34] X.Zhang, W.Hu, S.Maybank, X.Li, and M.Zhu. Sequential particle swarm optimization for visual tracking. in Proc. IEEE International Conference on Computer Vision and Pattern Recognition, Anchorage, USA, 2008
[35] Bradski G R, Clara S. Computer vision face tracking for use in a perceptual user interface [J]. Intel Technology Journal, 1998, 2 (2): 1215
[36] D.Comaniciu, V.Ranmesh and P.Meer. IEEE Transactions on Pattern Analysis and Machine Intelligence 25, 564(2003)
[37] URL http://en.wikipedia.org/wiki/HSV_color_space
[38] Kenneth R . Castleman著;朱志刚,林学訚,石定机等译.数字图像处理.北京:电子工业出版社,2000
[39] Ming-Hsuan Yang, David J.Kriegman, Narendra Ahuja. Detecting Faces in Images: A Survey. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2002, 24(1): 34-58
[40] Michael Oren, Constantine Papageorgiou, pawan Sinha, Edgar Osuna, Tomaso Poggio.Pedestrian detection Using Wavelet Templates. IEEE Internal Conference on Computer Vision and Recognition (CVPR). 1997, 00: 193
[41] P.Viola, M.J.Jones. Robust Real-Time Face Detection. International Journal of Computer Vision(IJCV). 2004, 57(2):137-154
[42] David Ross, Jongwoo Lim, Ruei-Sung Lin, Ming-Hsuan Yang. Incremental Learning for Robust Visual Tracking. In the International Journal of Computer Vsion, Special Issue:Learning for Vision. 2007
[43] Harris C, Stephens M.A. Combined Corner and Edge Detection. Proceedings of The Fourth Alvey Vision Conference .1988, 147-151
[44] Vincent Lepetit. Keypoints Recognition Using Randomized Trees. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2006, 28(9): 1465-1479
[45] Ozuysal Mustafa, Fua Pascal, Lepetit Vincent. Fast Keypoint Recognition in Ten Lines of Code. Computer Vision and Pattern Recognition, 2007 CVPR 07 IEEE Conference on.17-22 June 2007:1-8
[46] Dan Crisan, Arnaud Doucet. A survey of convergence results on particle filtering for practitioners, 2002
[47]盛骤等,概率论和数理统计.北京:高等教书出版社, 1989
[48] Greg Welch, Gary Bishop. An Introduction to the Kalman Filter. SIGGRAPH Course. 2001
[49]杨万海,多传感器数据融合及其应用.西安:西安电子科技大学出版社,2004
[50] Y Bar-Shalom and TE Formann. Tracking and Data Association.Academic Press, 1988
[51] T E Fortmann, Y Bar-Shalom and M Scheffe. Sonar tracking of multiple targets using joint probabilistic data association.IEEE Journal of Oceanic Engineering, vol.8, pp.173-184, 1983
[52] L Y Pao. Multi-sensor multi-target mixture reduction algorithms for target tracking. AIAA Journal of Guidance, Control and Dynamics, vol.17, pp.1205-1211, 1994
[53] D J Salmond. Mixture reduction algorithms for target tracking in clutter.Signal and Data Processing of Small Targets, SPIE 1305, O E Drummond, Ed, 1990, pp.434-445
[54] J. Vermaak, S. Godsill and P. P′erez. Monte Carlo filtering for multi-target tracking and data association. IEEE Transactions on Aerospace and Electronic systems, vol. 41,no. 1, pp. 309–332, 2005
[55] J.Lafferty, A.McCallum,F.Pereira. Conditional Random Fields:Probabilistic Models for Segmenting and Labeling Sequenee Data. In International Conference on Machine Learning,2001
[56] A.Quattoni,M.Collins,T.Darrell. Conditional random fields for object recognition.in Advances in Neural Information Processing Systems, 2004
[57] A.Quattoni, S.Wang, L-P.Morency, M.Collins,T.Darrell. Hidden-state Conditional Random Fields[J]. IEEE Transactions on Pattern Analysis and MachineIntelligence, 2007
[58] L.P.Morency, A.Quattoni, T.Darrell. Latent-Dynamic Discriminative Models for Continuous Gesture Recognition. in Proceedings IEEE Conference on Computer Vision and Pattern Recognition, 2007
[59] W.B.Wang, A.Quattoni, L.P.Morency, D.Demirdjian. Hidden Conditional Random Fields for Gesture Recognition. in IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2006, Washington, DC, USA
[60] A.Gunawardana, M.Mahajan, A.Acero, J.C.Platt. Hidden conditional random fields for phone classification. in International Conference on Speech Communication and Technology, 2005
[61] Y.P.Chu, X.Z.Ye, J.Qian, Y.Zhang, S.Y.Zhang. Adaptive foreground and shadow segmentation using hidden conditional random fields. Journal of Zhe Jiang University Science A, 2007, 8(4): 586~592
[62] Y Bar-Shalom and T E Fortmann. Tracking and Data Association. Academic Press, 1988.
[63] N. Wax. Signal-to-Noise Improvement and the Statistics of Tracking Populations. J. of Applied Physics, pp. 586-595, May 1955
[64] R. Sittler. An Optimal Data Association Problem in Surveillance Theory. IEEE Tran on Military Electronics, pp. 125-139, Apr. 1964
[65]Y. Bar-Shalom and E. Tse. Tracking in a Cluttered Environment with Probabilistic Data Association. Automatica, pp. 451-460, 1975
[66] R. Singer and J. Stein. An Optimal Tracking Filter for Processing Sensor Data of Imprecisely Determined Origin in Surveillance Systems. Proc.1971 IEEE CDC, FL, pp.171-175, 1971
[67] Blom H A P, Bar-Shalom Y. The Interacting Multiple Model Alogorithm for System with Marcovian Swithing Confficients. IEEE Trans. On AC, 1988
[68] http://web.engr.oregonstate.edu/~hess/downloads/hockey_track.mpg
[2] CollinsR.T.,etal . A system for Video Surveillance and Monitoring, CMU-RI-TR
[3] http://www-sop.inria.fr/orion/ADVISOR
[4] Matsuyama T. Cooperative Distributed Vision. In Proceedings of DARPA Image Understanding Workshop, 1998:365-384
[5] D.Comaniciu, V.Ramesh and P.Meer. Kernel-based object tracking. IEEE Transaction on Pattern Analysis and Machine Intelligence, 2003, 25(5): 564-577
[6] C.Yang, R.Duraiswami and L.Davis. Efficient mean-shift tracking via a new similarity measure. In:Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition, San Diego, CA, United States, 2005, I, 176-183
[7]彭宁嵩,杨杰,刘志等. Mean-Shift跟踪算法中核函数窗宽的自动选取.软件学报, 2005,16(9): 1542-1551
[8] J.Tu, H.Tao and T.Huang.Online updating appearance generative mixture model for meanshift tracking. In: Proceedings of Asian Conference on Computer Vision, Hyderabad,India, 2006, 1, 694-703
[9] Y.Chen,Y.Rui and T.S.Huang. JPDAF based HMM for real-time contour tracking. In: Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Kauai, Hawaii, 2001, 1, 543-550
[10] Xinyu Xu and Baoxin Li. Rao-Blackwellised Particle Filter for Tracking with Application in Visual Surveillance. Proceedings Joint IEEE International Workshop on VS-PETS, Beijing, October 15-18, 2005
[11] Swain, M.J, Ballard, D.H. Color indexing. International Journal of Computer Vision, 1991, 7(1):11~32
[12] Stricker, M., Orengo, M. Similarity of color images. In: Niblack, W.R., Rceds, J., eds. Proceedings of the SPIE2420, Storage and Retrieval for Image and Video DatabaseIII. San Jose, CA: SPIE, 1995. 381~392
[13] Liu, Zhong-wei. Color image retrieval using local accumulative histogram. Journal ofimage and graphics, 1998, 3(7): 533~537
[14] D. Androutsos, K.N.Plataniotis, A.N.Venetsanopoulos. A Novel vector-based approach to color image retrieval using a vector angular-based distance measure. Computer Vision and Image Understanding, 1999, 75(1/2): 46~58
[15] Pass G, Zabih R. Histogram refinement for content-based image retrieval. In: Proceedings of the 3rd IEEE Workshop on Applications of Computer Vision (WACV’96). Sarasota: IEEE Computer Society, 1996. 96~102
[18] Huang J. Image indexing using color correlograms. In: Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. San Juan: IEEE Computer Society, 1997. 762~768
[19] D.G.Lowe. Distinctive image features from scale-invariant keyPoints. IJCV, 60(2): 91-110 , 2004
[20] NUmmiaroK, Koller-Meier, Van GOOLL. Color features for tracking non-rigidobjeets,Special Issue on Visual Surveillanee, Chinese Journal of Automation,2003: 345-355
[21] Arulampalam M S, Marskell S, Gordon N, et al. A tutorial on particle filters for online nonlinear/non-Gaussian Bayesian tracking.IEEE Transactions On Signal Processing, 2002, 50(2): 174-188
[22] McGinnity S, Irwin G. Multiple model estimation using the bootstrap filter. IEEE Colloquium on Target Tracking and Data Fusion , 1998: 3/1-3/3
[23] Liu J.S and Chen R. Sequential Monte Carlo Methods for Dynamical Systems. Journal of the American Statistical Association, 1998, 93(5): 1032-1044
[24] Kitagawa G. Monte Carlo Filter and Smoother for Non-Gaussian Nonlinear State Space Models. Journal of Computational and Graphical Statistics, 1996, 5(1): 1-25
[25] Kennedy J, Eberhart R. Particle swarm optimization. Proceedings of the IEEE International Conference on Neural Network Piscataway, NJ, USA: IEEE, 1995: 1942– 1948
[26] R.Van Merwe, A.Doucet, N.de Freitas, et al. The unscented particle filter, Technical report CUED/F-INFENG/TR380, Cambridge University Engineering Department, August, 2000
[27] N.J.Gordon, D.J.Salmond and A.F.M.Smith. Novel approach to nonlinear/non-GaussianBayesian state estimation. IEEE Proceedings-F of Radar and Signal Processing, 1993, 140(2): 107-113
[28] T.Higuchi. Monte Carlo filter using the genetic algorithm operators.Journal of Statistical Computation and Simulation, 1997, 59(1): 1-23
[29] J.S.Liu and R.Chen. Sequential Monte Carlo methods for dynamic systems. Journal of the American Statistical Association, 1998, 93(443): 1032-1044
[30] G.Kitagawa. Monte carlo filter and smoother for non-gaussian nonlinear state space models. Journal of Computational and Graphical Statistics, 1996, 5: 1-25
[31] J.Kennedy and R.C.Eberhart. Particle swarm optimization , in Proc. IEEE Int’l Conf.on Neural Networks, pp.1942-1948, 1995
[32] M.Clerc and J.kennedy. The particle swarm-explosion, stability and convergence in a multidimensional complex space. IEEE Trans.on Evolutionary Computation , 6(1): 58-73, 2002
[33] M.P.Wachowiak, R.Smolikova and Y.Zheng. An approach to multimodal biomedical image registration utilizing particle swarm optimization. IEEE Trans.on Evolutionary Computation, 8(3): 289-301, 2004
[34] X.Zhang, W.Hu, S.Maybank, X.Li, and M.Zhu. Sequential particle swarm optimization for visual tracking. in Proc. IEEE International Conference on Computer Vision and Pattern Recognition, Anchorage, USA, 2008
[35] Bradski G R, Clara S. Computer vision face tracking for use in a perceptual user interface [J]. Intel Technology Journal, 1998, 2 (2): 1215
[36] D.Comaniciu, V.Ranmesh and P.Meer. IEEE Transactions on Pattern Analysis and Machine Intelligence 25, 564(2003)
[37] URL http://en.wikipedia.org/wiki/HSV_color_space
[38] Kenneth R . Castleman著;朱志刚,林学訚,石定机等译.数字图像处理.北京:电子工业出版社,2000
[39] Ming-Hsuan Yang, David J.Kriegman, Narendra Ahuja. Detecting Faces in Images: A Survey. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2002, 24(1): 34-58
[40] Michael Oren, Constantine Papageorgiou, pawan Sinha, Edgar Osuna, Tomaso Poggio.Pedestrian detection Using Wavelet Templates. IEEE Internal Conference on Computer Vision and Recognition (CVPR). 1997, 00: 193
[41] P.Viola, M.J.Jones. Robust Real-Time Face Detection. International Journal of Computer Vision(IJCV). 2004, 57(2):137-154
[42] David Ross, Jongwoo Lim, Ruei-Sung Lin, Ming-Hsuan Yang. Incremental Learning for Robust Visual Tracking. In the International Journal of Computer Vsion, Special Issue:Learning for Vision. 2007
[43] Harris C, Stephens M.A. Combined Corner and Edge Detection. Proceedings of The Fourth Alvey Vision Conference .1988, 147-151
[44] Vincent Lepetit. Keypoints Recognition Using Randomized Trees. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2006, 28(9): 1465-1479
[45] Ozuysal Mustafa, Fua Pascal, Lepetit Vincent. Fast Keypoint Recognition in Ten Lines of Code. Computer Vision and Pattern Recognition, 2007 CVPR 07 IEEE Conference on.17-22 June 2007:1-8
[46] Dan Crisan, Arnaud Doucet. A survey of convergence results on particle filtering for practitioners, 2002
[47]盛骤等,概率论和数理统计.北京:高等教书出版社, 1989
[48] Greg Welch, Gary Bishop. An Introduction to the Kalman Filter. SIGGRAPH Course. 2001
[49]杨万海,多传感器数据融合及其应用.西安:西安电子科技大学出版社,2004
[50] Y Bar-Shalom and TE Formann. Tracking and Data Association.Academic Press, 1988
[51] T E Fortmann, Y Bar-Shalom and M Scheffe. Sonar tracking of multiple targets using joint probabilistic data association.IEEE Journal of Oceanic Engineering, vol.8, pp.173-184, 1983
[52] L Y Pao. Multi-sensor multi-target mixture reduction algorithms for target tracking. AIAA Journal of Guidance, Control and Dynamics, vol.17, pp.1205-1211, 1994
[53] D J Salmond. Mixture reduction algorithms for target tracking in clutter.Signal and Data Processing of Small Targets, SPIE 1305, O E Drummond, Ed, 1990, pp.434-445
[54] J. Vermaak, S. Godsill and P. P′erez. Monte Carlo filtering for multi-target tracking and data association. IEEE Transactions on Aerospace and Electronic systems, vol. 41,no. 1, pp. 309–332, 2005
[55] J.Lafferty, A.McCallum,F.Pereira. Conditional Random Fields:Probabilistic Models for Segmenting and Labeling Sequenee Data. In International Conference on Machine Learning,2001
[56] A.Quattoni,M.Collins,T.Darrell. Conditional random fields for object recognition.in Advances in Neural Information Processing Systems, 2004
[57] A.Quattoni, S.Wang, L-P.Morency, M.Collins,T.Darrell. Hidden-state Conditional Random Fields[J]. IEEE Transactions on Pattern Analysis and MachineIntelligence, 2007
[58] L.P.Morency, A.Quattoni, T.Darrell. Latent-Dynamic Discriminative Models for Continuous Gesture Recognition. in Proceedings IEEE Conference on Computer Vision and Pattern Recognition, 2007
[59] W.B.Wang, A.Quattoni, L.P.Morency, D.Demirdjian. Hidden Conditional Random Fields for Gesture Recognition. in IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2006, Washington, DC, USA
[60] A.Gunawardana, M.Mahajan, A.Acero, J.C.Platt. Hidden conditional random fields for phone classification. in International Conference on Speech Communication and Technology, 2005
[61] Y.P.Chu, X.Z.Ye, J.Qian, Y.Zhang, S.Y.Zhang. Adaptive foreground and shadow segmentation using hidden conditional random fields. Journal of Zhe Jiang University Science A, 2007, 8(4): 586~592
[62] Y Bar-Shalom and T E Fortmann. Tracking and Data Association. Academic Press, 1988.
[63] N. Wax. Signal-to-Noise Improvement and the Statistics of Tracking Populations. J. of Applied Physics, pp. 586-595, May 1955
[64] R. Sittler. An Optimal Data Association Problem in Surveillance Theory. IEEE Tran on Military Electronics, pp. 125-139, Apr. 1964
[65]Y. Bar-Shalom and E. Tse. Tracking in a Cluttered Environment with Probabilistic Data Association. Automatica, pp. 451-460, 1975
[66] R. Singer and J. Stein. An Optimal Tracking Filter for Processing Sensor Data of Imprecisely Determined Origin in Surveillance Systems. Proc.1971 IEEE CDC, FL, pp.171-175, 1971
[67] Blom H A P, Bar-Shalom Y. The Interacting Multiple Model Alogorithm for System with Marcovian Swithing Confficients. IEEE Trans. On AC, 1988
[68] http://web.engr.oregonstate.edu/~hess/downloads/hockey_track.mpg