本論文提出了一種基於深度影像來實現快速且準確的手勢參數估測系統。此系統先將深度影像轉換為特徵向量，再以回歸函式得出手指關節參數。有別於目前手勢估測的兩種主要方向Model-based和Appearance-based，此方法只需要極短的時間便可計算出值域連續的參數。一開始先從深度影像中擷取出手部影像並標準化，透過隨機森林分類器找出手上的幾個特定點，將其轉換為不受手部旋轉影響的特徵向量之後，再使用預先訓練好的回歸函式計算出手勢參數。其中我們所使用的手部影像擷取系統可以不受手部旋轉影響而準確地定位出手掌中心，標準化使得手部的大小不隨距離遠近而變化。隨機森林分類器中所使用的分裂函式經過我們的修改後，得以處裡在影像平面上的旋轉情況，使其更加適合應用在手勢偵測上。實驗方面，我們將提出的方法所估算的參數與電子手套取得的實際參數做比對，來計算此系統的準確度。我們也統計出不同距離以及不同的手部旋轉角度對準確度的影響；並且，不同的測試者對系統的影響與其應對方法也在文章的最後一併討論。

This thesis presents a real-time and precise depth image based hand pose estimation method. The depth image obtained from Kinect is converted into a feature vector for regression functions to retrieve hand joint parameters. Different from the two mainly proposed methods, model-based and appearance-based, our approach retrieves continuous result within a short period of time. In the beginning, the hand region is segmented from the depth image. Some specific feature points on the hand are located by random forest classifier, and the relative displacements of these feature points would be transformed into a rotation invariant feature vector. Finally, the system retrieves the hand joint parameters by using regression functions which are trained off-line. The results of the proposed method are compared with the ground truth obtained by data glove to evaluate the system reliability. The effects of different distances and different rotation angles to the estimation accuracy are evaluated.

摘要
Abstract
Contents
List of Figures
1. Introduction
2. Related Work
2.1. Appearance-based Method
2.2. Model-based Method
3. Hand Feature Extraction
3.1. Hand Centroid Locating
3.2. Scale Normalization
4. Finger Joint Locating
4.1. Random Forest
4.2. Joint Point Extraction using Random Forest
5. Feature Vector and SVR
5.1. Regression Function with LIBSVM
5.2. Hand Model Rendering
6. Experimental Result
6.1. Hand Segmentation Result
6.2. Time Consumption
6.3. Reliability Evaluation
6.4. Accuracy at Different Distance
6.5. Rotated Gestures Testing
6.6. Application to Different Identity
6.7. Comparison with Other Approaches
7. Conclusion
Reference

[1] M. Półrola and A. Wojciechowski, "Real-Time Hand Pose Estimation Using Classifiers," ICCVG, pp. 573-580, September 2012.
[2] H. Teleb and G. Chang, "Data glove integration with 3D virtual environments," Systems and Informatics (ICSAI), pp. 107 - 112, May 2012.
[3] R. Y. Wang and J. Popović, "Real-Time Hand-Tracking with a Color Glove," ACM SIGGRAPH, August 2009.
[4] A. Erol, G. Bebis, M. Nicolescu, R. D. Boyle and X. Twombly, "Vision-based hand pose estimation: A review," CVIU, pp. 52-73, October 2007.
[5] J. Romero, H. Kjellström and D. Kragic, "Hands in action: real-time 3D reconstruction of hands in interaction with objects," Robotics and Automation (ICRA), pp. 458 - 463, May 2010.
[6] S. Miyamoto, T. Matsuo, N. Shimada and Y. Shirai, "Real-time and precise 3-D hand posture estimation based on classification tree trained with variations of appearances," ICPR, pp. 453 - 456, November 2012.
[7] M. d. L. Gorce, N. Paragios and D. J. Fleet, "Model-Based Hand Tracking with Texture, Shading and Self-occlusions," CVPR, pp. 1 - 8, June 2008.
[8] H. Hamer, K. Schindler, E. Koller-Meier and L. V. Gool, "Tracking a Hand Manipulating an Object," ICCV, pp. 1475 - 1482, October 2009.
[9] I. Oikonomidis, N. Kyriazis and A. A. Argyros, "Markerless and Efficient 26-DOF Hand Pose Recovery," Asian Conference on Computer Vision, pp. 744-757, 2010.
[10] I. Oikonomidis, N. Kyriazis and A. A. Argyros, "Efficient Model-based 3D Tracking of Hand Articulations using Kinect," British Machine Vision Conference, August 2011.
[11] J. Shotton, A. Fitzgibbon, M. Cook, T. Sharp, M. Finocchio, R. Moore, A. Kipman and A. Blake, "Real-time human pose recognition in parts from single depth images," CVPR, pp. 1297 - 1304, June 2011.
[12] H. Breu, J. Gil, D. Kirkpatrick and M. Werman, "Linear time Euclidean distance transform algorithms," PAMI, pp. 529 - 533, May 1995.
[13] L. Breiman, "Random Forests," Machine Learning, vol. 45, pp. 5-32, 1 October 2001.
[14] H. Drucker, C. J. C. Burges, L. Kaufman, A. Smola and V. Vapnik, "Support Vector Regression Machines," Advances in Neural Information Processing Systems 9, vol. 9, pp. 155-161, 1997.
[15] C. Cortes and V. Vapnik, "Support-Vector Networks," Machine Learning, vol. 20, pp. 273-297, September 1995.
[16] C. C. Chung and L. C. Jen, "LIBSVM : a library for support vector machines," ACM Transactions on Intelligent Systems and Technology, pp. 27:1--27:27, 2011.
[17] M. Bray, E. Koller-Meier and L. V. Gool, "Smart Particle Filtering for 3D Hand Tracking," Automatic Face and Gesture Recognition, pp. 675 - 680, May 2004.
[18] M. d. L. Gorce, D. J. Fleet and N. Paragios, "Model-Based 3D Hand Pose Estimation from Monocular Video," PAMI, pp. 1793 - 1805, September 2011.
[19] T. Tanimoto and K. Hoshino, "Real Time Posture Estimation of Human Hand for Robot Hand Interface," Universal Communication, pp. 303 - 308, December 2008.
[20] C. Keskin, F. Kiraç, Y. E. Kara and L. Akarun, "Real time hand pose estimation using depth sensors," ICCVW, pp. 1228 - 1234, November 2011.