在近年來人機介面的研究領域中，手勢辨識與重建一向是個相當重要的研究目標，使用手勢做為人機介面，有著高使用者親和力、自然與直觀等優勢。然而，在現今已實現的系統當中，大多數的系統都必須依靠使用者於手部附加額外的感測器或者標記以便系統進行參數擷取。如此一來，我們便必須負擔更高的建構成本，同時也會也會導致整個系統對於使用者而言並不友善。在這個狀況之下，我們嘗試去建立一個基於視覺的手部動作參數擷取系統，來增進使用者親合力以及降低成本。
在本篇論文中，我們建立了一個包含了結構限制以及運動學限制的三維手部模型，在使用該些限制之下，我們的手部模型運動方式，將會更為貼近真實的人體手部運動。在過去的研究中，預估類似手部動作參數這樣高維度參數空間的參數是極為耗時的，為了有效率的實現在高維度狀態空間中預估手部動作參數的系統，我們提出了 Separatable State Based Particle Filter (SSBPF)。這個方法將我們所要預估的參數切割為幾個不同的部份，然後以 Particle Filter 預估切割之後的參數空間中之參數。結合以上兩者，我們實現了手部動作參數擷取系統。在我們所提出的系統中，系統可以達到自動的進行初始化動作，而不需要由人工進行初始化。在準確率上，系統的平均擷取誤差在所使用的運算時間為每個畫面一秒時，可低於十五度。

In this paper, we form a 3-dimensional hand model with structural constraint and kinematical constraint. By using the constraints, our hand model behaviors more similar to ordinary human hand. In order to estimate human hand state in high degree of freedom space, we proposed separatable state based particle filter (SSBPF). This method tries to separate the states of the variable which we want to estimate into several different parts, and track each part with the particle filter. Combining the hand model and SSBPF, we implement a human hand gesture reconstruction system. Our system may initialize automatically and reconstruct human hand model with a 3D virtual hand in one second with less than 15 degree average error.

[1] A Imai, N Shimada, Y Shirai. “3-D hand posture recognition by training contour variation”. Proc. IEEE Int. Conf. Automatic Face and Gesture Recognition, 2004.
[2] V. Athitsos and S. Sclaroff. “Estimating 3D hand pose from a cluttered image”. IEEE Conf. on Computer Vision and Pattern Recognition, volume II, Madison, 2003.
[3] Imagawa K,Taniguchi R,Arita D, et al. “Appearance-Based recognition of hand shapes for sign language in low resolution image”. Asian Conference on Computer Vision, IEEE, 2000.
[4] Erik B. Sudderth, Michael I. Mandel, William T. Freeman, and Alan S. “Visual Hand Tracking Using Nonparametric Belief Propagation”. Computer Vision and Pattern Recognition Workshop, IEEE, 2004, Seoul.
[5] John Y. Lin, Ying Wu, Thomas S. Huang. “3D Model-Based Hand Tracking Using Stochastic Direct Search Method”. Proc. IEEE Int. Conf. Automatic Face and Gesture Recognition, 2004, Seoul.
[6] J Lee, TL Kunii, “Model-based Analysis Hand Posture”, Computer Graphics and Applications, IEEE, 1995
[7] James J. Kuch and Thomas S. Huang, “Vision Based Hand Modeling and Tracking
for Virtual Teleconferencing and Telecollaboration”, Proc. Int. Conf. Computer Vision, 1995
[8] S Kopf, T Haenselmann, W Effelsberg, “Shape-based Posture and Gesture Recognition in Videos”, Proc. SPIE, 2005
[9] B Moghaddam, A Pentland, “Maximum Likelihood Detection of faces and hands”, International Workshop on Automatic Face-and Gesture-Recognition, 1995.
[10] Y Wu, TS Huang, “View-independent Recognition of Hand Postures”, In Proc. Of IEEE Conf. on Computer Vision and Pattern Recognition, 2000
[11] J. Rehg and T. Kanade, “Model-based tracking of selfoccluding articulated objects.” In Proc. of IEEE Conf.Computer Vision, 1995.
[12] S. Lu, D. Metaxas, D. Samaras, and J. Oliensis, “Using multiple cues for hand tracking and model refinement”, In Proc. IEEE Int’l Conf. on Computer Vision, 2003.
[13] YI Abdel-Aziz, HM Karara, “Direct linear transformation from comparator coordinates into object space coordinates ”, Proceedings of the Symposium on Close-Range Photogrammetry, 1971.