基於機器學習於多深度攝影機架構下之即時籃球動作辨識

以作者查詢圖書館館藏

、以作者查詢臺灣博碩士

、以作者查詢全國書目

、勘誤回報

、線上人數：3

、訪客IP：3.15.25.183

姓名

郭瑋元(Wei-Yuan Kuo) 查詢紙本館藏

畢業系所

通訊工程學系

論文名稱

基於機器學習於多深度攝影機架構下之即時籃球動作辨識
(A Real-time Basketball Action Recognition based on Machine Learning Algorithm in Multi-View Environment)

相關論文

★ 基於區域權重之衛星影像超解析技術	★ 延伸曝光曲線線性特性之調適性高動態範圍影像融合演算法
★ 實現於RISC架構之H.264視訊編碼複雜度控制	★ 基於卷積遞迴神經網路之構音異常評估技術
★ 具有元學習分類權重轉移網路生成遮罩於少樣本圖像分割技術	★ 具有注意力機制之隱式表示於影像重建三維人體模型
★ 使用對抗式圖形神經網路之物件偵測張榮	★ 基於弱監督式學習可變形模型之三維人臉重建
★ 以非監督式表徵分離學習之邊緣運算裝置低延遲樂曲中人聲轉換架構	★ 基於序列至序列模型之 FMCW雷達估計人體姿勢
★ 基於多層次注意力機制之單目相機語意場景補全技術	★ 基於時序卷積網路之單FMCW雷達應用於非接觸式即時生命特徵監控
★ 視訊隨選網路上的視訊訊務描述與管理	★ 基於線性預測編碼及音框基頻週期同步之高品質語音變換技術
★ 基於藉語音再取樣萃取共振峰變化之聲調調整技術	★ 即時細緻可調性視訊在無線區域網路下之傳輸效率最佳化研究

檔案

[Endnote RIS 格式]

[Bibtex 格式]

[相關文章]

[文章引用]

[完整記錄]

[館藏目錄]

[檢視]

[下載]

本電子論文使用權限為同意立即開放。
已達開放權限電子全文僅授權使用者為學術研究之目的，進行個人非營利性質之檢索、閱讀、列印。
請遵守中華民國著作權法之相關規定，切勿任意重製、散佈、改作、轉貼、播送，以免觸法。

摘要(中)

人類動作辨識這項技術在電腦視覺及圖形識別等領域一直是重要的研究議題。這項技術可以廣泛應用在如醫學復健、運動動作訓練、電玩遊戲與監控系統等領域裡。在早期的動作辨識相關研究之所以不普及，主要是因為在設備的成本與使用空間需求上不是一般研究者可以負擔得起。但Kinect 體感攝影機的發布改變了這一現象，其低成本的優點使得大量研究者開始投入動作辨識的相關研究領域。
本論文利用 Kinect 裝置搭配由Windows公司推出的Kinect SDK2.0開發套件所提供的25個人體骨架關節點資訊與深度資訊製作出一套人體動作姿勢的辨識系統，用以解決昂貴的傳統深度攝影機所造成的設備成本問題並使用其中的骨架資訊來加強辨識準確率。本論文為了解決複雜的人體動作而產生的自我遮蔽問題和其他錯誤現象所造成的辨識率下降，故我們使用了兩台的Kinect攝影機來做到一個資料來源的互補，這樣即可以大幅降低因可能的遮蔽現象所造成的辨識誤判。之後我們再進行目標物的特徵擷取利用機器學習的方法建立的辨識系統，用以建立我們的資料庫。

摘要(英)

Human action recognition has been an important research in computer vision and computer graphics. It is widely used in entertainment, sports, medical applications and surveillance system. The traditional motion capture equipment is not usually affordable for normal developer. With the reasonable price of Kinect camera, low-cost human motion recognition becomes possible.
In this paper, we use multiple Kinect sensors and Kinect SDK as the tool to build our human action recognition system. This solves the problem of action recognition equipment costs. Using multiple Kinect cameras to solve the judging and correction error problems (such as self-occlusion and image noise...etc.) and using machine learning method to classified our features, it can make our recognition result with higher performance.
In our methods, we also have a detection of basketball to prevent that the subject is without ball, it makes our works more reasonable. Above of all, this paper have the action recognition rate to be more than 90% in real-time usage from three of the trained behaviors, i.e. right-hand dribble, left-hand dribble, and shooting behaviors.

關鍵字(中)

★ 人類動作辨識
★ 機器學習

關鍵字(英)

★ action recognition
★ multiple Kinects
★ Kinect SDK

論文目次

摘要 I
Abstract II
誌謝 III
目錄 IV
圖目錄 VI
表目錄 VIII
第一章緒論 1
1.1　研究背景 1
1.2　研究動機與目的 5
1.3　論文架構 6
第二章　相關研究環境介紹 7
2.1　Kinect體感攝影機 7
2.1.1　Kinect感測器之硬體規格 8
2.1.2　現有深度攝影機比較 9
2.1.3　Kinect軟體開發工具介紹 10
2.2 Kinect相關工具 13
2.2.1　Kinect Studio 13
2.2.2　Visual Gesture Builder 18
第三章　動作辨識相關技術介紹 19
3.1　動作辨識技術簡介 19
3.2　影像低階特徵擷取 20
3.2.1　色彩特徵 20
3.2.2　紋理特徵 21
3.2.3　形狀特徵 21
3.2.4　骨架特徵 22
3.2.5　深度特徵 23
3.3　機器學習之相關介紹 24
3.3.1　隨機森林 Random Forest 25
3.3.2　支持向量機 SVM 28
第四章提出之動作辨識技術 31
4.1　系統架構 31
4.2　VGB訓練 33
4.3　特徵擷取與訓練 35
4.3.1　VGB訊號 35
4.3.2　相對位置特徵 37
4.3.3　籃球輪廓偵測 39
4.3.4　特徵訓練 41
第五章實驗結果與分析討論 43
5.1　實驗環境介紹 43
5.2　實驗結果與討論 44
第六章　結論與未來展望 65
參考文獻 66

參考文獻

[1] A. Wenger; A. Gardner; C. Tchou; J. Unger; T. Hawkins; P. Debevec, "Performance relighting and reflectance transformation with time-multiplexed illumination," in ACM Transactions on Graphics (TOG), vol. 24, pp. 756-764, 2005.
[2] W. Li; Z. Zhang; Z. Liu, "Action recognition based on a bag of 3D points," in 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops (CVPRW),pp.9-14, 13-18, June 2010.
[3] K. Kaewplee; N. Khamsemanan; C. Nattee, "A rule-based approach for improving Kinect Skeletal Tracking system with an application on standard Muay Thai maneuvers," in 15th International Symposium on Soft Computing and Intelligent Systems (SCIS), 2014 Joint 7th International Conference on and Advanced Intelligent Systems (ISIS), vol., no., pp.281-285, 3-6b Dec. 2014.
[4] V. Tam and L. Ling-Shan, "Integrating the Kinect camera, gesture recognition and mobile devices for interactive discussion," Teaching, Assessment and Learning for Engineering (TALE), 2012 IEEE International Conference on, pp.H4C-11-H4C-13, Hong Kong, China, Aug. 2012.
[5] Ravikiran J, Kavi Mahesh, Suhas Mahishi, Dheeraj R, Sudheender S, Nitin V Pujari, "Finger Detection for Sign Language Recognition," Proceedings of the International MultiConference of Engineers and Computer Scientists, pp.489-493, Hong Kong, China, March 2009.
[6] C. Sun, T. Zhang, B. K. Bao, C. Xu, and T. Mei, "Discriminative Exemplar Coding for Sign Language Recognition with Kinect," Cybernetics, IEEE Transactions on, pp. 1-1, Switzerland, Lausanne, Jun 2013.
[7] H. Hui-Huang, T. Kang-Chun, C. Zixue, and H. Tongjun, "Posture Recognition with G-Sensors on Smart Phones," Network-Based Information Systems (NBiS), 2012 15th International Conference on, pp. 588-591, Australia ,Melbourne, Sept. 2012.
[8] J. E. Garrido, V. M. R. Penichet, M. D. Lozano, and L. A. Sanchez, "Mobility and memory training through movement interaction," Computer Science and Information Systems (FedCSIS), 2012 Federated Conference on, pp. 883-889, Wroclaw, Polska, Sept. 2012.
[9] A. Reiss and D. Stricker, "Introducing a modular activity monitoring system," Engineering in Medicine and Biology Society, EMBC, 2011 Annual International Conference of the IEEE, pp. 5621-5624, Boston, MA, Aug. 2011.
[10] Xbox360+Kinect 官方網站〔Online〕. Available:
http://www.xbox.com/zh-TW/kinect/
[11] R. Souvenir, A. Hajja, and S. Spurlock, "Gamesourcing to acquire labeled human pose estimation data," Computer Vision and Pattern Recognition Workshops (CVPRW), 2012 IEEE Computer Society Conference on, pp. 1-6, Providence, RI, June 2012.
[12] F. Lv; R. Nevatia, "Single view human action recognition using key pose matching and viterbi path searching," in 2007 IEEE Conference on Computer Vision and Pattern Recognition ( CVPR), 2007.
[13] H. Liu; L. Li, "Human Action Recognition Using Maximum Temporal Inter-Class Dissimilarity," in The Proceedings of the Second International Conference on Communications, Signal Processing, and Systems. Springer International Publishing, 2014.
[14] G. T. Papadopoulos; A. Axenopoulos; P. Daras, "Real-time skeleton-tracking-based human action recognition using kinect data," MultiMedia Modeling. Springer International Publishing, 2014.
[15] J. K. Aggarwal; L. Xia, "Human activity recognition from 3d data: A review," Pattern Recognition Letters, pp. 70-80, 2014.
[16] H. Li; M. Greenspan, "Multi-Scale Gesture Recognition from Time-Varying Contours," Proc. IEEE Int’l Conf. Computer Vision, vol. 1, pp. 236-243, 2005.

[17] A. Efros; A. Berg; G. Mori; J. Malik, "Recognizing Action at a Distance," in Proc. IEEE Int’l Conf. Computer Vision, vol. 2, pp. 726- 733, 2003
[18] A. Elgammal; V. Shet; Y. Yacoob; L.S. Davis, "Learning Dynamics for Exemplar-Based Gesture Recognition," in Proc. IEEE Conf. Computer Vision and Pattern Recognition, vol. 1, pp. 571-578, 2003.
[19] C. Schuldt; I. Laptev; B. Caputo, "Recognizing Human Actions: A Local SVM Approach," Proc. Int’l Conf. Pattern Recognition, vol. 3, pp. 32-36, 2004.
[20] L. Gorelick; M. Blank; E. Shechtman; M. Irani; R. Basri, "Action as Space-Time Shapes," Appeared first in the Tenth IEEE International Conference on Computer Vision (ICCV), 2005
[21] Suraj Ho, ”A motion capture framework to encourage correct execution of sport exercises”, Master thesis presented to the Faculty of Sciences at the VU University Amsterdam.
[22] C.H.Hsu, J.M.Chen, “Human Motion Capture and Mistake Correction Using Kinect Camera”, Master thesis of science department of computer science and engineering tatung university, July. 2014
[23] OpenNI, “OpenNI User Guide, “ available at: https://github.com/OpenNI/OpenNI/blob/master/Documentation/OpenNI_UserGuide.pdf, Dec. 2015.
[24] Heresy’s, “Kinect for Windows的錄影工具: Kinect Studio”, https://kheresy.wordpress.com/2015/08/10/data-recording-and-playback-tool-kinect-studio/, Aug. 2015.
[25] Microsoft, “Visual Gesture Builder: A Data-Driven Solution to Gesture Detection”, Published for the Xbox One XDK, Sep. 2014.
[26] J. Yamato, J. Ohya, and K. Ishii, “Recognizing Human Action in Time-Sequential Images using Hidden Markov Model,”In proc. IEEE CVPR,pp. 377-385, 1992.
[27] I. Haritaoglu, D. Harwood, and L. S. Davis, “W4: Real-Time Surveillance of People and Their Activities, “IEEE Trans. On Pattern Analysis and Machine Intelligence, vol. 22, no. 8,pp. 809-830, 2000.
[28] M. Ferman, A. M. Tekalp, and R. Mehrotra, “Robust Color Histogram Descriptors for Video Segment Retrieval and Identification”, IEEE Transactions on Image Processing, vol. 11, no. 5, pp. 497-508, May 2002.
[29] H. Tamura, S. Mori, T. Yamawaki, “Texture features corresponding to visual perception”, IEEE Transactions on Systems, Man and Cybernetics, vol. 8, no. 6, pp.460-473, 1978.
[30] L. Breiman., Random forests. Mach. Learning, 45(1):5-32, 2001.
[31] Heresy’s, “K4W v2 C++ Part 7:偵測、追蹤人體骨架”, https://kheresy.wordpress.com/2015/03/03/k4w-v2-part-7-user-skeleton/ , Mar. 2015.
[32] C. Liu, Y. Kong, X. Wu and Y. Jia, “Action recognition with discriminative mid-level features,” in Pattern Recognition (ICPR), 2012 21st International Conference on, 11-15 Nov. 2012.
[33] T. Harada, T. Mori, T. Sato, “Human Posture Probability Density Estimation Based on Actual Motion Measurement and Eigenpostures,” IEEE International Conference on Systems, Man and Cybernetics, vol. 2, no.10-13, pp.1595-1600, 2004.
[34] C. H. Hsia, C. H. Chien, H. W. Hsu, Y. F. Chang and J. S. Chiang, "Analyses of basketball player field goal shooting postures for player motion correction using kinect sensor," Intelligent Signal Processing and Communication Systems (ISPACS), 2014 International Symposium on, Kuching, 2014, pp. 222-225.
[35] Z. Liu, Z. Miao and Y. Huo, "A realtime human action recognition method based on single view key poses in sports video," 6th International Conference on Wireless, Mobile and Multi-Media (ICWMMN 2015), Beijing, 2015.
[36] C. H. Chung, C. J. Lin, “LIBSVM: A Library for Support Vector Machines, “ACM Transactions on Intelligent Systems and Technology (TIST), 2011 Article No. 27.

指導教授

張寶基(Pao-Chi Chang)

審核日期

2016-11-8

推文