跳到主要內容

簡易檢索 / 詳目顯示

回結果列表
研究生： 穆罕德
Mohammad Firdaus Ardiansyah
論文名稱： 瑜珈姿態辨識-使用多重KINECT
Multiple Kinects Motion Sensing for Yoga
指導教授： 葉士青
Yeh Shih-Ching
口試委員:
學位類別： 碩士
Master
系所名稱： 資訊電機學院 - 資訊工程學系
Department of Computer Science & Information Engineering
論文出版年： 2013
畢業學年度： 101
語文別： 英文
論文頁數： 39
外文關鍵詞： Yoga
相關次數： 點閱：8  下載：0 
分享至:
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報

    • ABSTRACT
    Human motion tracking is receiving increasing attention from researchers of different fields of study nowadays. The interest is motivated by a wide range of applications, such as wireless healthcare, surveillance, and human-computer interaction. A complete model of human consists of both the movements and the shape of the human body skeleton. There are so many ways to track human motion, from marker sensing, marker less, optical, etc. Kinect sensor developed by Microsoft had done human tracking in 3d perspective and widely used in both in commercial and experiments purpose. However this Kinect method tracking by estimating depth image yield good result for simple gestures. In this study a novel approach is proposed to track complicated movement such as yoga movement with three Kinects in order to overcome kinect’s limitation.

    摘 要
    人體動作偵測最近在不同領域的研究中正逐漸變成一個令人關注的議題。無線健康照護、監控系統、人機互動等應用為促成此篇論文的動機。一個完整的人體模型包含動作以及骨架的形狀。 有許多方法可以用來追蹤人體動作，配戴感測器、光學追蹤等。微軟開發的kinect可以在三維空間中完成人體姿態辨識，而且在商業、實驗用途都被廣泛地使用。然而kinect藉由估計深度圖像來追蹤的方法只對簡單的姿勢才能有較好的結果。本論文提出了一個新的方法，利用三部kinect克服單部kinect的限制，使其可以追蹤如瑜珈般複雜的動作。

    NATIONAL CENTRAL UNIVERSITY iii DEPARTMENT OF COMPUTER SCIENCE AND iii INFORMATION ENGINEERING iii 摘 要 vi ABSTRACT vii Keywords : Human body tracking, Kinect, Multiple Kinect, Yoga Motion vii TABLE OF CONTENT viii LIST OF TABLE x LIST OF FIGURE xi ACKNOWLEDGEMENT xii CHAPTER 1: INTRODUCTION 1 1.1. Background and Motivation 1 1.2. Research Object 2 1.3. Structure of the thesis 2 CHAPTER 2: LITERATURE REVIEW 3 2.1. Motion Tracking with Depth Camera (Kinect) 3 2.2. Toolkit for Kinect 5 2.2.1. OpenNI 5 2.2.2. Microsoft Kinect SDK 6 2.2.4. Comparation from Toolkits 8 2.3. Motion Tracking with Multiple Depth Cameras (Kinects) 9 2.4. Tracking Algorithm 11 CHAPTER 3: METHODOLOGY 12 3.1. Research design 12 3.2. System design 13 3.2.1. System Environment 13 3.2.2. Training system 13 3.2.3. Single Kinect System 16 3.2.4. Multiple Kinect System 17 3.2.5. Assessment system 20 3.3. Experiment Design 21 3.4. Measurement and Analysis 24 3.4.1. Performance assessment 24 3.4.2. Human perception assessment 24 CHAPTER 4: DATA ANALYSIS AND RESULTS 25 4.1. Performance Assessment. 25 4.2. Analysis and Measurement Human Behavior and Perception 28 CHAPTER 5: CONCLUSION 30 5.1. Conclusion 30 5.2. Future Work 30 References 31 Appendix 33 Questionnaire 35

    References
    1. Zhang, S.X., W.;Gong, J.;Yin,Y., A Novel Human Motion Tracking Approach Based on a Wireless Sensor Network. International Journal of Distributed Sensor Networks, 2013. Volume 2013, Article ID 636052.
    2. Organicmotion, What is Motion Capture? url http://www.organicmotion.com/products/openstage/motion-capture (accessed July 19 2013), 2013.
    3. Knies, R., Academics, Enthusiasts to Get Kinect SDK. url http://research.microsoft.com/en-us/news/features/kinectforwindowssdk-022111.aspx (accessed 2013-07-20), 2011.
    4. Microsoft, Human Interface Guideline. Microsoft Kinect SDK Documentation, 2013.
    5. Destructoid, The Limitations of Kinect. url http://www.destructoid.com/blogs/pneuma08/the-limitations-of-kinect-186724.phtml (accessed 2013-07-20), 2010.
    6. Microsoft, Kinect. url: http://www.xbox.com/en-us/kinect (accessed May 27, 2013), 2010.
    7. Webb, J. and J. Ashley, Beginning kinect programming with the microsoft kinect SDK. The expert's voice in Microsoft. 2012, New York: Apress. xvii, 306 p.
    8. Cong, R.W., R., How It Works: Xbox Kinect. url: http://www.jameco.com/Jameco/workshop/howitworks/xboxkinect.html (accessed on june 9th 2013).
    9. NI, O., OPEN NI. url: http://www.openni.org/ (accessed May 27, 2013), 2013.
    10. PrimeSense, NiTE Middleware. url: http://www.primesense.com/solutions/nite-middleware/ (accessed june 9th 2013), 2013.
    11. Microsoft, Kinect for Windows Feature. url http://www.microsoft.com/en-us/kinectforwindows/Discover/Features.aspx (accessed 2013-07-20), 2013.
    12. Will, H., Kinect for Windows SDK beta vs OpenNI. url: http://labs.vectorform.com/2011/06/windows-kinect-sdk-vs-openni-2/ (accessed June 9th, 2013), 2011.
    13. Cai, Q.A., J. K.;, Tracking Human Motion Using Multiple Cameras. IEEE, 1996. Proceedings of ICPR(1996).
    14. Dockstader, S.T., M.;, Multiple Camera Tracking of Interacting and Occluded Human Motion. PROCEEDINGS OF THE IEEE, 2001. 89(2001).
    15. Ian, W., Performance and Motion Capture Using Multiple Xbox Kinects. Rochester Institute of Technology, 2013, 2013.
    16. Shotton, J., et al., Real-Time Human Pose Recognition in Parts from Single Depth Images. Communications of the Acm, 2013. 56(1): p. 116-124.
    17. Moll, G.P.B., A.; Helten, T.; Muller, M.; Seidel, H.P.; Rosenhahn, B., Multisensor-Fusion for 3D Full-Body Human Motion Capture. IEEE, 2010.
    18. Rosenhahn, B.K., R.; Metaxas, D., Human motion - understanding, modelling, capture and animation. Netherlands: Springer, 2008.
    19. Gall, J., et al., Optimization and Filtering for Human Motion Capture. International Journal of Computer Vision, 2010. 87(1-2): p. 75-92.
    20. Obdrzalek, S.K., G.; Ofli, F.; Bajcsy, R., Accuracy and Robustness of Kinect Pose Estimation in the Context of Coaching of Elderly Population. Annual International Conference of the IEEE EMBS, 2012.
    21. Li, R., et al., 3D Human Motion Tracking with a Coordinated Mixture of Factor Analyzers. International Journal of Computer Vision, 2010. 87(1-2): p. 170-190.
    22. Georgakopoulos, D.Y., W.Q., Circuit noise reduction by analogue lowpass filtering and data averaging. IEEE Electronics Letters Online No: 20010782, 2001.

    QR CODE
    :::