人體運動的自動分析為近年來很熱門且極具有挑戰性的研究主題，但是，絕大部分的分析數據及研究成果都是設計在2D影像資訊上。因此，提出了一個新的研究方向，針對3D的人體運動分析為主軸，透過VICON Motion Capture System (動態捕捉系統)所輸出的純3D的檔案"CSM"，作為實驗分析的資料依據，本論文所設計的搜尋系統，提供給使用者手動調整權重值，可以動態調整至使用者想要看的結果，而這套系統將人體身上的關節點分為身體、頭部、手、腳...等。
本論文的研究方法是比對動作軌跡，利用補點運算，將三維檔案內的移動軌跡修正、再將其轉換成VRML格式之3D動畫運動，資料庫中大約分為四大類運動：排球、壘球、網球，與羽毛球，根據這些動作粹取出具特徵的關鍵動作，依據分類結果，再將關鍵動作加入權重值，接著比對關鍵動作之間的差異性並將其差異量化，透過數據資料可找出相似動作並以相似度做排序。

Automatic analysis of motions is an interesting and challenge research topic. But, most existing motion analysis systems are designed based on 2D video information. This thesis proposes a 3D motion retrieval system which supports the efficient analysis of 3D motions of sports made by VICON motion capture system to substitute for the traditional 2D information. The developed system provides adaptive weights function which dynamically calculated according to user’s perception of motion features. Besides, the representation of human skeleton includes head, hip, knee, elbow, wrist, etc. further to aggregate important features in sport motions based on kinematics is proposed. Finally, with the motion capture of video technology, it is possible to automatically represent, analyze and adjust the 3D motions of user via Internet.

第一章 緒論	1
1.1 研究動機	1
1.2 研究目的	2
1.3 相關背景	3
1.4 系統架構	13
1.5 論文組織	14
第二章 相關研究	16
2.1 國內相關研究	16
2.2 國外相關研究	18
第三章 3D動作資料庫之建立	24
3.1 VICON SYSTEM拍攝與製作	24
3.2 CSM檔案校正轉換	27
3.3 3ds Max骨骼之建立	28
3.4 VRML轉換	30
3.5 動作資料之分類	33
第四章 3D運動特徵值之表示法	36
4.1 軌跡表示法	36
4.2 角度表示法	38
4.3 動量表示法	41
第五章 3D動作特徵之比對	43
5.1 比對單一物體之移動軌跡	43
5.2 軌跡補點演算法	50
5.3 比較兩人體運動之移動軌跡	53
第六章 系統實作與結果	57
6.1 軌跡系統設計之開發程式	57
6.2 3D物件瀏覽器	60
6.3 系統介面	64
6.4 實驗結果	70
第七章 結論與未來研究方向	73
7.1 結論與貢獻	73
7.2 未來的期望與研究方向	74
參考文獻	77
附錄 		80
圖 目 錄
圖一：VICON 攝影機....................................... 4
圖二：3ds Max 匯入3D 專用軌跡檔案........................ 5
圖三：利用 parallelgraphics 公司之VRML plug-in 開啟之檔案8
圖四：VRMLPAD 介面...................................... 10
圖五：此圖顯示出一顆球的行進軌跡路徑.................... 11
圖六：系統架構圖........................................ 13
圖七：受試者黏貼上感光式接收球.......................... 25
圖八：受試著擺出T-Pose 姿勢............................. 25
圖九：VICON 人體骨架圖.................................. 26
圖十：羽球運動軌跡圖.................................... 27
圖十一：CSM 檔案內容.................................... 28
圖十二：3Ds character 所提供之人體骨骼.................. 29
圖十三：綠色圈選區為扭曲變形的人體骨骼.................. 30
圖十四：人體骨架圖...................................... 31
圖十五：拍攝初始狀態(左)，前手刺拳完整人體模型(右上) .... 33
與其轉換後精簡MODEL(右下) .............................. 33
圖十六：角度表示法...................................... 39
圖十七：軌跡資料庫...................................... 44
圖十八：DTW 軌跡比對結果................................ 45
圖十九：DTW 演算法...................................... 46
圖二十：Detail Threshold 實驗數據圖..................... 48
圖二十一：Rough Threshold 搜尋結果圖.................... 50
圖二十二：軌跡原始圖.................................... 51
圖二十三：缺少軌跡點圖.................................. 51
圖二十四：補點過後圖(1.2 倍) ............................ 52
圖二十五：補點過後圖(1.3 倍) ............................ 53
圖二十六：權重值搜尋結果圖.............................. 55
圖二十七：Borland JBuilder 的環境....................... 58
圖二十八：Microsoft Visual C#.NET 環境說明.............. 59
圖二十九：Cortona Browser Plug-in 初始介面.............. 61
圖三十：3D Motion Retrieval System 程式初始畫面......... 64
圖三十一：透過IE Browser 預覽VRML 檔案.................. 64
圖三十二：3D Motion Retrieval System 搜尋結果........... 66
圖三十三：詳細時間點的運動停留.......................... 67
圖三十四：放大鏡鈕觀看不同角度之運動.................... 68
圖三十五：資料庫檢索.................................... 69
圖三十六：cortona plug-in 按鍵說明...................... 70
圖三十七：壘球(內野)側投法之搜尋結果.................... 71
圖三十八：壘球(內野)側投法之Precision/Recall 圖......... 72
表 目 錄
表一：VRML 歷史與其版本演進.............................. 6
表二：支援VRML 之Plug-In 比較........................... 8

[1].http://www.vicon.com/
[2].http://www.autodesk.com/3dsmax
[3].R. Carey, G. Bell and C. Marrin. “The Virtual Reality Modeling Language”. Apr. 1997, ISO/IEC DIS 14772-1. [Online]:http://www.web3d.org/Specifications/.
[4].J. Hartman and J. Wernecke. The VRML 2.0 Handbook: Building Moving Worlds on the Web, Addison Wesley, 1996.
[5].嚴子翔. “VRML虛擬實境網頁語言”, 知城數位科技股份有限公司, ISBN 957-0435-48-8。
[6].Dejan V. and D. Saupe. "3D Model Retrieval", Proceedings of Spring Conference on Computer Graphics 2000, Comenius University Press, Bratislava, Slovakia, pp. 89-93, May 2000.
[7].Vranic, D. V., ”Content-based Search for 3D-objects” in Proceeding of Computational Intelligence and Multimedia Applications. Fourth International Conference on ICCIMA, pp. 266 –270, 2001.
[8].C. Zhang and T. Chen. “Indexing and Retrieval of 3D Models Aided by Active Learning” Proceedings of the ninth ACM international conference on Multimedia, Ottawa, Canada, pp. 615 – 616, 2001.
[9].T. K. Shih, Yi-Ping Hung, and Jiun-Hung Huang. “Content-Based Information Retrieval in Large VR Scene Database” Second International Workshop on Intelligent Multimedia Computing and Networking(IMMCN’2002) 
[10].M. Heczko, D. Keim, D. Saupe, and D. V. Vranic. “A Method for Similarity Search of 3D objects”(in German),Proc. Of BTW 2002,Oldenburg,Germany, pp. 384-401,March 2001.
[11].M. Novotni and R. Klein. “A Geometric Approach to 3D Object Comparison  , “ Proc. of Shape Modelling International (SMI 2001), Genova, Italy, 2001, (to appear).
[12].C. Zhang and T. Chen. “Active Learning for Information Retrieval: Using 3D Models as An Example", Carnegie Mellon Technical Report: AMP01-04.
[13].C. Zhang and T. Chen. "Efficient Feature Extraction for 2D/3D Objects in Mesh Representation", ICIP 2001,Thessaloniki,Greece, Oct. 7-10, 2001.
[14].Go Nishimura. et al. ”3D Object Retrieval based on Shape Similarity”, IEICE database system 122-62,pp479-485,2000(In Japanese)
[15].Y. Emoto et al., ”A Feature for Shape-based Retrieval of 3D Database from Geometric Information”, IPSI Symposium Series Vol.200 No 17.ISSN 1344-0640, pp155-162, 2000(In Japanese)
[16].D.V. Vranic and D. Saupe. “3D Model Retrieval.” Proc. Of Spring Conference on Computer Graphics and Applications(SCCG2000), Budmerice Manor, Slovakia,pp. 89-93,May 2000.
[17].Motofumi T. Suzuki, Toshikazu Kato and Nobuyuki Otsu. “A Similarity Retrieval of 3D Polygonal Models Using Rotation Invariant Shape Descriptors”, Systems, Man, and Cybernetics, 2000 IEEE International Conference on, pp. 2946-2952, Volume 4, 2000.
[18].M. T. Suzuki and T. kato,H. Tsukune. “A Statistical Method For Subjective Retrieval of 3D objects”, Trans. IEICE,D-I, VOL. J82-D-I,pp184-192.1999.
[19].J. Yi, D. Chelberg. “Model-based 3D Object Recognition Using Baycsian”, Computer Vision and Image Understanding. Vol.69, No.1,1998,pp.87-105.
[20].Eric Paquet and Marc Rioux. “Nefertiti: A Query by Content Software for Three-dimensional Models Databases Management”, 3-D Digital Imaging and Modeling, 1997.Proceedings. International Conference on Recent Advances in, pp. 345-352, 1997.
[21].T. P. Wallace and P. A. Wintz. “An Efficient Three-dimensional Aircraft Recognition Algorithm using Normalized Fourier Fescriptors”, Computer Graph. Image Process., vol. 13, pp. 99–126, 1980.
[22].Wen-Yen Chang. “Motion Capture Data Analysis and Segmentation”, 清華大學碩士論文
[23].Shih-Pin Chao, Chih-Yi Chiu, Jui-Hsiang Chao, Shi-Nine Yang and Tsang-Ken Lin. “Motion Retrieval and Its Application to Motion Synthesis”, International Conference on Distributed Computing Systems Workshops 2004
[24].Timothy K. Shih, Ching-Hseng Wang, Yuan-Kai Chiu, Yi-Tsou Hsin and Chun-Hong Huang . “On Automatic Actions Retrieval of Martial Arts” , IEEE Conference On Multimedia and Expo 2004
[25].Takeshi Yabe and Katsumi Tanaka. “Similarity Retrieval of Human Motion as Multi Stream Time Series Data” , Database Applications in Non-Traditional Environments, 1999
[26].Feng Liu, Yueting Zhuang, Fei Wu, and Yunhe Pan. “3D Motion Retrieval with Motion Index Tree” , Computer Vision and Image Understanding
[27].Lucas Kovar	and Michael Gleicher. “Automated Extraction and Parameterization of Motions in Large Data Sets” ACM Transactions on Graphics (TOG), Volume 23 , Issue 3,  2004
[28].YiWang, Gang Qian and Tanassis Rikakis. “RobustPause Detection Using 3D Motion Capture Data”, IEEE International Conference on Acoustics, Speech, and Signal Processing 2005
[29].Ying Piao, Kazutaka Hayakawa and Jun Sato. “Space-Time Invariants for Recognizing 3D Motions from Arbitrary Viewpoints under Perspective Projection” , Third International Conference on Image and Graphics, 2004.
[30].Taketoshi Mori, Yushi Segawa, Masamichi Shimosaka and Tomomasa Sato. “Hierarchical Recognition of Daily Human Actions based on Continuous Hidden Markov Models” , Sixth IEEE International Conference on  Automatic Face and Gesture Recognition, 2004.