Skip navigation

DSpace

機構典藏 DSpace 系統致力於保存各式數位資料(如:文字、圖片、PDF)並使其易於取用。

點此認識 DSpace
DSpace logo
English
中文
  • 瀏覽論文
    • 校院系所
    • 出版年
    • 作者
    • 標題
    • 關鍵字
    • 指導教授
  • 搜尋 TDR
  • 授權 Q&A
    • 我的頁面
    • 接受 E-mail 通知
    • 編輯個人資料
  1. NTU Theses and Dissertations Repository
  2. 電機資訊學院
  3. 電信工程學研究所
請用此 Handle URI 來引用此文件: http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/66074
完整後設資料紀錄
DC 欄位值語言
dc.contributor.advisor鄭士康(Shyh-Kang Jeng)
dc.contributor.authorPo-Chi Huangen
dc.contributor.author黃柏齊zh_TW
dc.date.accessioned2021-06-17T00:20:57Z-
dc.date.available2022-06-04
dc.date.copyright2012-06-29
dc.date.issued2012
dc.date.submitted2012-06-21
dc.identifier.citation[1] J. Aggarwal and Q. Cai, “Human Motion Analysis: A Review,” Computer Vision and Image Understanding, vol. 73, no. 3, pp. 428–440, 1999.
[2] C. W. Chu, O. C. Jenkins, and M. J. Mataric, “Markerless kinematic model and motion capture from volume sequences,” IEEE Conf. Computer Vision and Pattern Recognition, Madison, WI, vol. 2, pp. 475–482, June 2003.
[3] http://www.xsens.com
[4] A. Baak, M. Muller, G. Bharaj, H. P Seidel, and C. Theobalt, 'A data-driven approach for real-time full body pose reconstruction from a depth camera,' IEEE International Conference on Computer Vision, pp. 1092–1099, November 2011.
[5] 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, “IEEE Conf. Computer Vision and Pattern Recognition, pp. 1297–1304, June 2011.
[6] S. Hussmann, T. Ringbeck, and B. Hagebeuker, “A performance review of 3D TOF vision systems in comparison to stereo vision systems,” in Stereo Vision, pp. 103–120, 2008.
[7] T. B. Thomas Moeslund, H. Adrian, and K. Volker, “A survey of advances in vision-based human motion capture and analysis,” Computer Vision and Image Understanding, vol. 104, no. 2–3, pp. 90–126, November 2006.
[8] Y. Mao, W. Xianwang, Y. Ruigang, R. Liu, and M. Pollefeys, 'Accurate 3D pose estimation from a single depth image,' IEEE International Conference on Computer Vision, pp. 731–738, November 2011.
[9] C. Plagemann, V. Ganapathi, D. Koller, and S. Thrun, 'Real-time identification and localization of body parts from depth images,' IEEE International Conference on Robotics and Automation, pp. 3108–3113, May 2010.
[10] V. Ganapathi, C. Plagemann, D. Koller, and S. Thrun, “Real time motion capture using a single time-of-flight camera,” IEEE Conference on Computer Vision and Pattern Recognition, pp. 755–762, June 2010.
[11] D. Grest, J. Woetzel, and R. Koch. “Nonlinear body pose estimation from depth images,” Annual meeting of the German Association for Pattern Recognition, pp. 285–292, August 2005.
[12] P. J. Besl, N. D. McKay, 'A Method for Registration of 3-D Shapes,' IEEE Transactions on Pattern Analysis and Machine Intelligence, pp. 239–256, February 1992.
[13] M. Y. Qiao, J. Cheng;, and W. C. Zhao, “Model-based Human Pose Estimation with Hierarchical ICP from single Depth Images,” Advances in Automation and Robotics, Vol, vol. 2, pp. 27–35, Dubai, December 2011.
[14] M. Siddiqui and G. Medioni, 'Human pose estimation from a single view point, real-time range sensor,' IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshop, pp. 1–8, June 2010.
[15] Y. Zhu, B. Dariush, and K. Fujimura, “Controlled human pose estimation from depth image streams,” IEEE Proc. CVPR Workshop on TOF Computer Vision, June 2008.
[16] C. C. Yang. “Human pose estimation using depth map and particle swarm optimization” M.A. thesis, Nation Taiwan University, Taiwan R.O.C., 2012.
[17] Z. Chen, 'Bayesian filtering: From Kalman filters to particle filters, and beyond,' Adaptive Systems Laboratory, McMaster University, Tech. Rep., 2003.
[18] M. S. Arulampalam, S. Maskell, N. Gordon, and T. Clapp, 'A tutorial on particle filters for online nonlinear/non-Gaussian Bayesian tracking', IEEE Trans. Signal Processing, vol. 50, pp. 174–188, 2002.
[19] A. Bissacco, A. Chiuso, and S. Soatto, 'Classification and Recognition of Dynamical Models: The Role of Phase, Independent Components, Kernels and Optimal Transport,' IEEE Transactions on Pattern Analysis and Machine Intelligence, vol.29, no.11, pp. 1958–1972, November 2007.
[20] G. Bradski and A. Kaehler, “Learning OpenCV,” O’Reilly Media, Inc., 2008.
[21] Y. Zhu, B. Dariush, and K. Fujimura. “Kinematic Self Retargeting: A framework for human pose estimation,” Journal of Computer Vision and Image Understanding, vol. 114, no. 12, pp. 1362–1375, December 2010.
dc.identifier.urihttp://tdr.lib.ntu.edu.tw/jspui/handle/123456789/66074-
dc.description.abstract利用單一相機辨識影像中人體的姿態一直是電腦視覺中一個相當熱門的研究議題,同時具有很多的應用,例如擴增實境、人機介面、電玩遊戲、醫療應用與智慧監控系統。然而,只利用單一相機辨識人體姿態有著許多困難,例如人有很多複雜的姿勢表現以及隨機沒有規律的動作,或是每個動作中身體部位間互相的遮蔽,都會造成單相機人體姿勢辨識系統產生錯誤的辨識結果。
在本研究中,我們提出一個以人體模型為基礎的人體姿態辨識方法。我們比對該模型與深度影像,找出最符合輸入影像的人體模型,進而得到人體真正的姿勢並且辨識人體各個身體部位 (例如: 軀幹、肩膀、手肘、…等)。 為了降低比對過程中得計算複雜度,我們採用階層式比對方法: 先比對出軀幹部分,接著再利用比對出的軀幹資訊找出四肢的部位。 除此之外,為了有效追蹤所偵測出來的身體部位,我們採用粒子濾波器演算法進行人體動作追蹤。該演算法可以有效追蹤隨機運動的物體並且不需要龐大的計算資源。與其他的方法相比較,我們的方法不需要大量的訓練資料與龐大的資料庫,也不需要去尋找一些瑣碎又不通用的特徵點。結果顯示,本研究所提出之人體姿勢辨識方法具有高計算效率以及高精準度,並可以有效追蹤人體之隨機動作與辨識相互遮蔽的人體部位。本系統之追蹤誤差僅0.06公尺,標準差則為0.04公尺,而運行速度則約每秒鐘20幀,使用的運算平台為Intel Core i3-2100 CPU且不需利用圖形處理器進行加速。
zh_TW
dc.description.abstractRecognizing human body poses is a challenging task because of varied human poses and unpredictable human movement. To address these problems, we propose a model-based approach for human body pose recognition from a single-view depth camera. The proposed algorithm applies an articulated cylinder model to detect human pose and track them based on a particle filter without numerous training data or heuristic detectors. To reduce high degrees of freedom, we adopt a hierarchical method that detects torso and limbs successively. Moreover, we take the advantage of a particle filter to track complex human motion and the results show that the proposed system is robust in human motion tracking. The qualitative evaluation shows that our method can deal with self-occlusion problem and ambiguous human motion effectively, and the quantitative evaluation shows that the average tracking error is 0.06 meters with a standard deviation of 0.04 meters. The proposed method tracks human poses successfully at the speed of 20 frames per second on a laptop with Intel Core i3-2100 CPU and without graphic processing unit.en
dc.description.provenanceMade available in DSpace on 2021-06-17T00:20:57Z (GMT). No. of bitstreams: 1
ntu-101-R99942121-1.pdf: 7051716 bytes, checksum: 14fbd462e37e7bc3727e26325e62d5d3 (MD5)
Previous issue date: 2012
en
dc.description.tableofcontents口試委員會審定書 #
誌謝 i
中文摘要 ii
ABSTRACT iii
CONTENTS iv
LIST OF FIGURES vi
LIST OF TABLES vii
Chapter 1 Introduction 1
1.1 Human Body Pose Recognition 1
1.2 Problem Statement 1
1.3 Depth Measurement Techniques 2
1.4 Contributions 3
1.5 Chapter Outline 4
Chapter 2 Literature Review 6
2.1 Learning-Based Methods for Human Pose Recognition 6
2.2 Model-Based Methods for Human Pose Recognition 7
2.3 Summary 8
Chapter 3 Background 10
3.1 Recursive Bayesian Filtering Framework 10
3.2 Particle Filter 11
Chapter 4 Methodology 13
4.1 System Overview 13
4.2 Human Body Model 14
4.3 Implementation of Particle Filter 17
4.3.1 Initialization 18
4.3.2 State Prediction 18
4.3.3 Weight Calculation 19
4.3.4 Resampling and State Estimation 20
Chapter 5 Results and Discussions 21
5.1 Experimental Environment 21
5.2 Quantitative Evaluation 21
5.2.1 Euclidean Distance Evaluation 22
5.2.2 Success Rate Evaluation 23
5.3 Qualitative Evaluation 25
5.4 Limitations of the Proposed System 28
Chapter 6 Conclusions 29
REFERENCES 30
dc.language.isoen
dc.subject人體動作追蹤zh_TW
dc.subject人體姿勢辨識zh_TW
dc.subject粒子濾波器zh_TW
dc.subjectparticle filteren
dc.subjecthuman body pose recognitionen
dc.subjecthuman motion trackingen
dc.subjectdepth image analysisen
dc.subjecthuman-computer interactionen
dc.title利用單深度相機之人體姿勢辨識系統zh_TW
dc.titleHuman Body Pose Recognition from a Single-View Depth Cameraen
dc.typeThesis
dc.date.schoolyear100-2
dc.description.degree碩士
dc.contributor.oralexamcommittee丁建均(Jian-Jiun Ding),傅楸善(Chiou-Shann Fuh)
dc.subject.keyword人體姿勢辨識,粒子濾波器,人體動作追蹤,zh_TW
dc.subject.keywordhuman body pose recognition,human motion tracking,particle filter,human-computer interaction,depth image analysis,en
dc.relation.page32
dc.rights.note有償授權
dc.date.accepted2012-06-21
dc.contributor.author-college電機資訊學院zh_TW
dc.contributor.author-dept電信工程學研究所zh_TW
顯示於系所單位:電信工程學研究所

文件中的檔案:
檔案 大小格式 
ntu-101-1.pdf
  未授權公開取用
6.89 MBAdobe PDF
顯示文件簡單紀錄


系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。

社群連結
聯絡資訊
10617臺北市大安區羅斯福路四段1號
No.1 Sec.4, Roosevelt Rd., Taipei, Taiwan, R.O.C. 106
Tel: (02)33662353
Email: ntuetds@ntu.edu.tw
意見箱
相關連結
館藏目錄
國內圖書館整合查詢 MetaCat
臺大學術典藏 NTU Scholars
臺大圖書館數位典藏館
本站聲明
© NTU Library All Rights Reserved