傳統的人體行為分析通常是利用身體的外形為基礎，根據人體進行不同行為時身體外形會隨著行為變化之特性，利用不同的描述方式描述人體的各種姿勢，進而利用其進行行為的分析與辨識。因此，行為辨識率的好壞與人體姿勢的描述的方式有很大的關係。
本論文提出一種新的人體姿勢描述方式，有別於傳統的人體外形描述方式透過主動式外觀模型（Active Appearance Models）的特性，根據外形資訊與紋理資訊建立出各種人體姿勢模組，分別以多組特徵點及其所形成的路徑描述各種人體姿勢的外形。利用主動式外觀模型所建立之姿勢模組能夠更精確的找出姿勢模組相對應的人體姿勢，在透過姿勢模組比對的機制即可從姿勢模組搜尋的結果中找出與前景物最相似的姿勢。因此，此前景物之姿勢即可決定；之後，再透過姿勢模組搜尋所得到的姿勢組合利用建立好人體行為之隱藏馬可夫模型（Hidden Markov Models）計算出前景物行為於不同行為模組之機率值進行行為的分析與辨識。
實驗結果證明，主動式外觀模型應用於人體姿勢的模組化是可行的，而且對於姿勢的搜尋與判定皆有很好的正確率；此外，利用前景物姿勢的組合透過隱藏馬可夫模型進行行為的辨識的確能夠辨識出前景物之行為。

In this thesis, a human body posture classification is proposed based on the Active Appearance Model（AAM）. It is a model-driven technique for modeling various human body postures. Basically, two features, a shape and a texture features, were extracted from the segmented blobs. In constructing the posture models, a set of annotated samples were collected for training the posture models. The interested points are then defined and segmented to represent the body shapes. Whenever an image is inputted, the interest points are searched and matched with the constructed models using the AAM searching algorithm. Finally, the best matched posture is selected from the constructed models. Basically, human motion is composed of a sequence of discrete postures. The discrete postures can be identified by the proposed AAM matching algorithm. They are the corresponding motion strings in Hidden Markov Models (HMM). After the human motion models have been built, an input motion can be recognized by matching the motion string with the highest probability on each HMM motion model. Experiments were conducted and the results were illustrated to demonstrate the performance of the proposed method. The AAM techniques can be employed to successfully identify human postures and high accuracy rates can be achieved. Moreover, human behavior can be accurately identified by combining the posture detection and HMM.

[1] D. M. Gavrila, “The Visual Analysis of Human Movement: A survey,” Computer Vision and Image Understanding, vol. 73, no.1, pp. 82 - 98, 1999.
[2] J. K. Aggarwal and Q. Cai, “Human Motion Analysis: a Review,” Computer Vision and Image Understanding, vol. 73, no.3, pp. 428 - 440, 1999.
[3] S. Ju, M. Black, and Y. Yacoob, “Cardboard People: A Parameterized Model of Articulated Image Motion,” International Conference on Face and Gesture Analysis, 1996.
[4] Fujiyoshi and A. J. Lipton, “Real-Time Human Motion Analysis by Image Skeletonization,” IEEE workshop on Application of Computer Vision, pp.15 - 21, 1998.
[5] I. Haritaoglu, D. Harwood, and L. S. Davis, “Ghost: A Human Body Part Labeling System Using Silhouettes,” International Conference on Pattern Recognition, pp. 77 – 82, 1998.
[6] C. Wren, A. Azarbayejani, T. Darrell, and A. Pentland, “Pfinder: Real-Time Tracking of the Human Body,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 19, No. 7, pp. 780-785, July 1997.
[7] C. C. Chen, J. W. Hsieh, Y. T. Hsu, and C. Y. Huang, “Segmentation of Human Body Parts Using Deformable Triangulation,” International Conference on Pattern Recognition, vol. 1, pp.355-358, 2006.
[8] J. W. Davis and A. F. Bobick, “The Recognition of Human Movement Using Temporal Templates,” IEEE Transaction on Pattern Analysis and Machine Intelligence, vol. 23 no.3, pp. 257-267, 2001.
[9] D. Y. Chen, H. Y. Mark Liao, H. R. Tyan, and C. W. Lin, “Automatic Key Posture Selection for Human Behavior Analysis,” IEEE International Workshop on Multimedia Signal Processing, Shanhai, China, November 2005.
[10] I. Haritaoglu, D. Harwood, and L. S. Davis, ”W4：Who? When? Where? What? A Real-Time System for Detecting and Tracking People,” International Conference on Face and Gesture Recognition, April, 14-16, 1998.
[11] J. Barron, D. Fleet, and S. Beauchemin, “Performance of Optical Flow Techniques,” International Journal of Computer Vision, vol. 12, no. 1, pp. 42-77, 1994.
[12] C. Anderson, P. Burt, and G. van der Wall, “Change Detection and Tracking Using Pyramid Transformation Techniques,” In Proceedings of SPIE - Intelligent Robots and Computer Vision, vol. 579, pp. 72-78, 1985.
[13] T. Horprasert, D. Harwood, and L. S. Davis, “A Statistical Approach for Real-time Robust Background Subtraction and Shadow Detection,” IEEE International Conference on Computer Vision, Frame-Rate Workshop, pp.1-19, 1999.
[14] T. F. Cootes, G. J. Edwards, and C. J. Taylor, “Active Appearance Models,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 23, no. 6, pp. 681-685, Jun., 2001.
[15] L. R. Rabiner, “A Tutorial on Hidden Markov Models and Applications in Speech Recognition,” Proceeding of the IEEE, pp. 257-286, 1989.
[16] S. Eickeler, A. Kosmala, and G. Rigoll, “Hidden Markov Model Based Continuous Online Gesture Recognition,” International Conference on Pattern Recognition, Brisbane, vol. 2, pp.1206-1208, 1998.
[17] Y. Iwai, H. Shimizu, and M. Yachida, “Real-Time Context-based Gesture Recognition using HMM and Automaton,” IEEE International Workshop on Recognition, Analysis and Tracking of Faces and Gestures in Real-time Systems, Kerkyra, Greece, pp. 127-134, 1999.
[18] N. Liu and B. C. Lovell, “Gesture Classification Using Hidden Markov Models and Viterbi Path Counting,” Digital Image Computing: Techniques and Applications, Sydney, pp 273-282, 2003.
[19] L. R. Rabiner, “A Tutorial on Hidden Markov Models and Selected Applications in Speech Recognition,” IEEE, vol. 77, no. 2, pp.257-286, Feb. 1989.
[20] M. B. Stegmann, B. K. Ersbøll, and R. Larsen, “FAME - A Flexible Appearance Modelling Environment,” IEEE Transactions on Medical Imaging, vol. 22(10), 2003.