隨著多媒體產業的快速發展，人機介面（Human Computer Interface）也越來越受到重視，各種電腦相關產業紛紛跳脫以往以平面的設備操控多媒體，改成以更直覺的三維輸入設備來操控，例如：遊戲產業的Wii控制器（Wii Remote）以及能感測深度資訊的Kinect等，此等透過使用者的肢體來當作輸入設備可以更方便讓使用者與多媒體溝通，降低使用者學習操作多媒體的時間，所以利用手勢來控制多媒體也越來越受到重視。
因此，本論文主要是提供一種手勢追蹤的方法，利用具有深度資訊的攝影機，透過一些影像處理技術以及三維的運動估計，達成手勢辨識與追蹤，讓使用者可以利用手勢來控制多媒體系統。由於手勢的自由度很高，為了簡化手勢的問題，目前是採取手部剛體的追蹤，主要是追蹤手腕的三維旋轉角度以及三維的平移量。

Recently, Human Computer Interface has attracted more and more attention. Instead of using traditional mice, keyboards and joysticks, nowadays many computer-related industries have adopted more intuitive three-dimensional input devices to control multimedia. Wii Remote and Kinect are two such examples which allow users to more easily communicate with the devices/computers. Such input devices also reduce the efforts that a user has to spend on learning how to interact with the multimedia devices. Therefore, using hand gestures to control the media has become more important.
In this paper, we propose a hand gesture tracking system based on a camera with depth information, together with a few image processing techniques and three-dimensional motion estimation methods, to achieve gesture recognition and tracking, so that users can use gestures to control a multimedia system. In order to simplify the problem of high degrees of freedom associated with hand gestures, currently we are only tracking the rigid motions of a hand, namely the three-dimensional rotations and displacements.

[1] Thayananthan A., Stenger B., Torr P.H.S., and Cipolla R.“Model-Based Hand Trackin Using a Hierarchical Bayesian Filter.” PAMI, 2006.
[2] Shimada N., Kimura K., and Shirai Y. “Real-Time 3-D Hand Posture Estimation based on 2-D Appearance Retrieval Using Monocular Camera.” ICCV, 2001.
[3] Thayananthan A., Stenger B., Torr P.H.S., and Cipolla R. “Shape Context and Chamfer Matching in Cluttered Scenes.” CVPR, 2003.
[4] Stenger B. “Template-Based Hand Pose Recognition Using Multiple Cues.” ACCV, 2006.
[5] Dhawale P., Masoodian M., and Rogers B. “Bare-hand 3D gesture input to interactive systems.” CHINZ, 2006.
[6] Chik D., Trumpf J., and Schraudolph N.N. “3D Hand Tracking in a Stochastic Approximation Setting.” ICCV 2007.
[7] Wang R. Y. and Popovi′c J. “Real-Time Hand-Tracking with a Color Glove.” SIGGRAPH, 2009.
[8] Lee T. and Höllerer T. “Handy AR: Markerless Inspection of Augmented Reality Objects Using Fingertip Tracking.” ISWC, 2007.
[9] Lu S., Metaxas D., Samaras D., and Oliensis J. “Using Multiple Cues for Hand Tracking and Model Refinement.” CVPR, 2003.
[10] Riccardo Pompeo web site. http://www.sixtyfourbit.org/index.html.
[11] 黃俊捷 “互動雙足式機器人之設計與實現(I)-手勢辨識” 2008
[12] 三維運動估計PDF檔。 http://robolab.sjtu.edu.cn/home/files/7-2_0.pdf