本論文中，提出一個使用空間以及時間資訊進行不同攝影機之間的行人辨識系統，可以判斷經過兩攝影機間的行人是否為同一人，系統是以單一攝影機移動行人偵測以及追蹤所得到的資訊做為辨識依據，並且攝影機視野之間不必須存在交集。我們的方法中以亮度轉換函式之機率主成分分析建立兩攝影機間的亮度關係，並使用局部色彩直方圖交集衡量兩行人外觀之相似度形成兩行人之觀察機率，兩攝影機之間時間與空間的關係是以進入/離開區域之高斯混和模型做為基礎，建立區域間行人行進時間之高斯模型做為區域間轉移機率，並整合進入/離開區域高斯混合模型以及區域間轉移機率，形成行人在攝影機間之轉移機率模型，用以衡量兩行人在攝影機間的轉移機率。本論文所提出的方法可以解決兩攝影機之間的亮度差異以及有效降低攝影機拍攝行人角度的差異所造成辨識上的困難，所提出的轉移機率衡量方式也可以減少因為進入/離開區域模型建立結果之正確性影響，而造成之誤判。

In this thesis, we present a people identification system across different cameras using spatial and temporal information. With the system, we could decide whether the person observed by each camera associated with the same person. The identification mechanism of this system is based on the information obtained from inter-camera motion people detection and tracking. In our method, cameras are not necessarily overlapping. We learn the lighting relationships among cameras by probabilistic principal component analysis subspace of brightness transfer function and using local color histogram intersection for observation probability creation between people observed by each cameras. The spatio-temporal relationship is based on gaussian mixture model of entry/exit zone. We build the gaussian model of inter-camera traveling time as the model of transition probability between two entry/exit zones. The transition probability that the person transfer from one camera to another camera can be estimated by the model transition probability between cameras combined gaussian mixture model of entry/exit zone and transition probability between entry/exit zones. The system is able to handle the situation that the lighting condition is different between each cameras and minimize the pose variation of people between two cameras. Our transition probability measurement is robust to the model of entry/exit zone without highly accuracy.

[1] A. Alahi, P. Vandergheynst, M. Bierlaire and M. Kunt, "Cascade of Descriptors to Detect and Track Objects Across Any Network of Cameras", Submitted to Computer Vision and Image Understanding, 2009
[2] J. Annesley, V. Leung, A. Colombo, J. Orwell and S. A. Velastin, "Fusion of Multiple Features for Identity Estimation", In Proceedings of the IEE International Conference on Imaging for Crime Detection and Prevention, 2006
[3] H. Bay, T. Tuytelaars and L. V. Gool, "Surf: Speeded Up Robust Features", Proceedings of the European Conference on Computer Vision, 2006
[4] K. W. Chen, C. C. Lai, Y. P. Hung and C. S. Chen, "An Adaptive Learning Method for Target Tracking across Multiple Cameras", Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 1-8, 2008.
[5] K. W. Chen, P. J. Lee and Y.P. Hung, "Egocentric View Transition for Video Monitoring in a Distributed Camera Network", Advances in Multimedia Modeling, 2011
[6] Q. Cai and J. K. Aggarwal, "Tracking Human Motion in Structured Environments Using a Distributed-Camera System”, IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 2, No. 11, pp. 1241-1247, Nov 1999
[7] R. Collins, A. Lipton, H. Fujiyoshi and T. Kanade, "Algorithms for Cooperative Multi-Sensor Surveillance", Proceedings of the IEEE, vol. 89, no. 10, pp. 1456–1477, October 2001.
[8] W. R. Ding, H. G. Li, Z. Jiang and X. J. Li, "Unsupervised Spatio-Temporal Multi-Human Detection and Recognition in Complex Scene", International Congress on Image and Signal Processing, 2009
[9] M. Farenzena, L. Bazzani, A. Perina, V. Murino, and M. Cristani, "Person Re-Identification by Symmetry-Driven Accumulation of Local Features", Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2010
[10] D. Gray and H. Tao, "Viewpoint Invariant Pedestrian Recognition with an Ensemble of Localized Features", European Conference on Computer Vision, 2008
[11] N. Gkalelis, A. Tefas and I. Pitas, "Human Identification From Human Movements", International Conference on Image Processing, 2009
[12] Y. Guo, Y. Shan, H. Sawhney and R. Kumar, "Peet: Prototype Embedding and Embedding Transition for Matching Vehicles over Disparate Viewpoints", Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2007
[13] T. Huang and S. Russell, "Object Identification: A Bayesian Analysis with Application to Traffic Surveillance", Artificial Intelligence, vol. 103, no. 1-2, pp. 77–93, Aug. 1998.
[14] A. Itai and H. Yasukawa, "Personal Identification Using Footstep Based on Wavelets", Intelligent Signal Processing and Communications, 2006
[15] O. Javed, Z. Rasheed, O. Alatas and M. Shah, "Knight: A Real Time Surveillance System for Multiple Overlapping and Non-Overlapping Cameras", In International Conference on Multimedia and Expo, July 2003
[16] O. Javed, Z. Rasheed, K. Shafique and M. Shah, "Tracking across Multiple Cameras with Disjoint Views", In The Ninth IEEE International Conference on Computer Vision, 2003
[17] O. Javed, K. Shafique and M. Shah, "Appearance Modeling for Tracking in Multiple Non-overlapping Cameras”, Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2005.
[18] S. Khan, O. Javed, Z. Rasheed and M. Shah, "Human Tracking in Multiple Cameras", In The Eighth IEEE International Conference on Computer Vision, July 2001.
[19] S. Khan and M. Shah, "Consistent Labeling of Tracked Objects in Multiple Cameras with Overlapping Fields of View", IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 25, No. 10, pp.1355-1360, October 2003
[20] V. Kettnaker and R. Zabih, "Counting People from Multiple Cameras", In IEEE International Conference on Multimedia Computing and Systems, Vol. 2, pp. 267–272, 1999
[21] D. Lowe, "Distinctive Image Features from Scale-Invariant Keypoints", International Journal of Computer Vision, Vol. 60, No. 2, pp. 91-110, 2004
[22] D. Makris, T. Ellis and J. Black, "Bridging the Gaps between Cameras", Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2004
[23] F. Porikli, "Inter-Camera Color Calibration using Cross-Correlation　Model Function", In IEEE Int. Conf. on Image Processing, 2003.
[24] U. Park, A. Jain, I. Kitahara, K. Kogure and N. Hagita, "ViSE: Visual Search Engine Using Multiple Networked Cameras", International Conference on Pattern Recognition, 2006
[25] B. Stanciulescu, O. Hamdoun, F. Moutarde and B. Steux, "Interest Points Harvesting in Video Sequences for Efficient Person Identification", The Eighth International Workshop on Visual Surveillance, 2008
[26] B. Song and A. Roy-Chowdhury, "Stochastic Adaptive Tracking In A Camera Network, International Conference on Computer Vision, 2007
[27] C. Stauffer, "Learning to Track Objects through Unobserved Regions", In Proceedings of IEEE Workshop on Motion, 2005
[28] C. Siebler, K. Bernardin and R. Stiefelhagen, "Adaptive Color Transformation for Person Re-identification in Camera Networks.pdf", Proceedings of the Fourth ACM/IEEE International Conference on Distributed Smart Cameras, 2010
[29] C. Stauffer and K. Tieu, "Automated Multi-Camera Planar Tracking Correspondence Modeling", Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2003
[30] D. Scholeicher and L. M. Bergasa, "People Tracking and Recognition using the Multi-Object Particle Filter Algorithm and Hierarchical PCA Method", The International Conference on "Computer as a tool", 2005
[31] Y. Shan, H. S. Sawhney and R. Kumar, "Unsupervised Learning of Discriminative Edge Measures for Vehicle Matching between Non-Overlapping Cameras", IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol.30, No.4, pp. 700-711, 2008
[32] K. Tieu, G. Dalley and W. Grimson, "Inference of Non-Overlapping Camera Network Topology by Measuring Statistical Dependence", Proceedings of the IEEE International Conference on Computer Vision, 2005.
[33] M. E. Tipping and C. M. Bishop, "Probabilistic Principal Component Analysis", Journal of the Royal Statistical Society, Series B, Vol. 61, No. 3, pp. 611–622, 1999.
[34] D. J. Wang, C. H. Chen, T. Y. Chen and C. T. Lee, "People Recognition for Entering & Leaving a Video Surveillance Area", International Conference on Innovative Computing, Information and Control, 2009
[35] X. Wang, G. Doretto, T. Sebastian, J. Rittscher and P. Tu, "Shape and Appearance Context Modeling", International Conference on Computer Vision, 2007