在建立能互動的虛擬實境、智慧型的人機介面以及逼真的電腦動畫時，人體動作的擷取是非常重要的一部分。階層式的骨架模型更是被廣泛的應用，以連接在一起的線段表現人體的骨架和關節。在近年的研究中，很多研究學者試著以合乎經濟成本的設備估算出骨架中的資訊。我們提出一個不需貼上標記而可以重建人體關節骨架資訊的方法。此方法結合了近年來在分析統計領域被大量應用的技術-等構映圖(ISOMAP)，以處理人體軀幹和肢體間可能相互遮蔽的問題。這樣的解法是基於一個假設，人體動作分布的非線性特質是因為人體關節的旋轉，而實驗的結果也的確如此。透過高斯混合模型，我們可以對套用ISOMAP後的結果作出分段，將頭、手、腳的部位分離出來，並且找到其端點。利用以上的資訊，配合我們所產生出來、相當於肌肉的樣本點，我們藉著找到最大重疊，將事先所定義的人體骨架擬合到重建出來的體積空間。最後我們以實例驗證本方法之有效性。

The acquisition of human motion data is essential for creating interactive virtual environments, intelligent user interfaces, and computer animations. Hierarchical skeleton models are widely used to represent virtual human whose body is composed of chains of bones with interconnecting joints. In recent studies, many researchers try to estimate the parameters of the skeleton with commercial equipments. In this study, we present a markerless motion capture approach for articulated kinematic structure. This approach utilizes the manifold learning technology, ISOMAP, to deal with the situation when the occlusion of limbs and body occurs. It is based on the assumption that the nonlinearity of rigid-body kinematic motion is introduced by rotations about the joint axes. By applying Gaussian Mixture Model on the result of ISOMAP, we can easily segment each part of body, such as head, hands, legs and torso. Then we can find the end points of limbs and head. With all the information above, we fit a pre-defined human body skeleton onto the constructed volume by finding the maximum overlap between the volume and the sample points which we generated as muscle. Several experimental examples are given to demonstrate the effectiveness of proposed approach.

[1] C. Theobalt, M. Magnor, P. Schueler, H.P. Seidel, Combining 2D Feature Tracking and Volume Reconstruction for Online Video-Based Human Motion Capture . in Proceedings of Pacific Graphics 2002, Beijing, China. p.96-103.

[2] C. Theobalt, M. Magnor, P. Schueler, H.P. Seidel, Multi-Layer Skeleton Fitting for Online Human Motion Capture. in Proceedings of 7th International Fall Workshop on Vision, Modeling and Visualization (VMV 2002), Erlangen, Germany. p.471-478.

[3] M.-C. Silaghi, R.Plaenkers, R. Boulic, P. Fua, and D. Thalmann. Local and global skeleton fitting techniques for optical motion capture. In Modeling and Motion Capture Techniques for Visual Environments, number 1537 in LNAI, NO 1537, pages 26-40. Springer, 1998.

[4] D. Gavrila. The visual analysis of human movement. CVIU, 73(1):82-98, January, 1999.

[5] C. Wren, A. Azarbayejani, T. Darrell, and A. Pentland. Pfinder: Real-time tracking of the human body. PAMI, 19(7):780-785, 1997.

[6] I. Mikic, M. Trivedi, E. Hunter, P. Cosman Human Body Model Acquisition and Tracking Using Voxel Data. In International Journal of Computer Vision, 53(3), 199-233, 2003.

[7] K. Cheung, T. Kanade, J.-Y. Bouguet, and M. Holler. A real time system for robust 3D voxel reconstruction of human motions, In Proc. of CVPR, volume 2, pages 714-720, June 2000.

[8] E. de Aguiar, C. Theobalt, M. Magnor, H. Theisel, H.-P. Seidel, Marker-free Model Reconstruction and Motion Tracking from 3D Voxel Data, Proc. IEEE Pacific Graphics (PG'04), Seoul, South Korea, pp.101-110, October 2004.

[9] Stephen J. McKenna, Kenny Morrison. A comparison of skin history and trajectory-based representation schemes for the recognition of user-specified gestures. Pattern Recognition 37(5): 999-1009 (2004).

[10] Chi-Wei Chu, Odest Chadwicke Jenkins, Jaja J. Mataric. Markerless Kinematic Model and Motion Capture from Voluem Sequences. IEEE Computer Vision and Pattern Recognition(CVPR 2003), pp. II: 475-482, 2003.

[11] C. Theobalt, E. de Aguiar, M. Magnor, H. Theisel, H.-P. Seidel, Marker-free Kinematic Skeleton Estimation from Sequences of Volume Data. Proc. of ACM Symposium on Virtual Reality Software and Technology (VRST), p.57-64, Hong Kong, China, 2004.

[12] I. Cohen and H. Li, Inference of Human Postures by Classification of 3D Human Body Shape IEEE International Workshop on Analysis and Modeling of Faces and Gestures, in conjunction with ICCV'03.

[13] I. Cohen, H. Li and M.W. Lee. Human Body Posture Inference for Immersive Interaction, ACM, International Workshop on Immersive Telepresence , ACM Multimedia 2002, Juan Les Pins, France. 6 December, 2002.

[14] Rudolf Mester, Til Aach, Lutz Dümbgen: Illumination-Invariant Change Detection Using a Statistical Colinearity Criterion. DAGM-Symposium 2001: 170-177.

[15] G. Slabaugh, W. B. Culbertson, T. Malzbender, and R. Shafer. A survey of methods for volumetric scene reconstruction from photographs. International Workshop on Volume Graphics 2001, Stoney Brook, New York, June 21-22, 2001.

[16] Wojciech Matusik, Chris Buehler, Ramesh Raskar, Steven J. Gortler, Leonard McMillan. Image-based visual hulls. SIGGRAPH 2000: 369-374.

[17] Laurentini, A. The Visual Hull Concept for Silhouette Based Image Understanding. IEEE PAMI 16,2 (1994), 150-162.

[18] K.M. Cheung, T. Kanade, J. Bouguet, and M. Holler. A Real Time System for Robust 3D Voxel Reconstruction of Human Motions, in Proceedings of IEEE Conference on Computer Vision and Pattern Recognition 2000 (CVPR '00), Vol. 2, pages 714-720, Hilton Head Island SC, June 2000.

[19] German K. M. Cheung, Simon Baker, Jessica K. Hodgins, Takeo Kanade: Markerless Human Motion Transfer. 3DPVT 2004: 373-378.

[20] J.B. Tenenbaum, V. de Silva and J. C. Langford. A global geometric framework for nonlinear dimensionality reduction. Science, vol. 290, pp. 2319--2323, 2000.