近年來，隨著低成本的深度攝影機問世，藉此直接擷取到包含人體關節座標的資料稱之為體感資料，而分析高維度之體感資料時往往需耗費大量的計算時間，使得這個問題具有相當的挑戰性；在許多替代方案中，類神經網路當中的自組織映射圖被證實在處理體感資料上有不錯的效果，而其特色在於透過訓練後的自組織映射圖，可將一串體感資料序列容易且精確的對應成一串動作索引；但是，自組織映射圖的訓練過程中相當耗時且過程繁瑣。有鑑於此，本研究採用了結合階層式架構至自組織映射圖的方法當中；本研究的方法又稱作生長四元樹自組織映射圖，目的在於幫助我們有效減少訓練時的計算複雜度，並且同時保留了自組織映射圖技術的特色。最後，我們採用了一個WorkoutSU-10 的測試資料集來進行實驗評估，結果顯示我們所提出的方法對於檢索出相似的動作序列有不錯的效率和顯著的檢索效果。

With low-cost depth cameras are released recently, motion data containing 3D coordinates of skeleton joints during a time period can be directly captured. Nevertheless, analyzing the motion data is usually a challenging problem and requires huge computation costs because of the high-dimensionality. Among several alternatives, the self-organizing map (SOM) is verified to be an effective technique to handle such motion data. Specifically, a captured motion sequence can be easily and precisely mapped to form an indexed motion string through the use of a trained SOM. However, the training process of the SOM is high computation complexity and is thus typically tedious. In view of this, we propose in this work to incorporate a hierarchical structure into the SOM technique. Generally speaking, our approach named as GQSOM (growing quadtree self-organizing map) helps to significantly reduce the required computation complexity while preserving the effectiveness of the SOM technique. Empirical studies using the WorkoutSU-10 exercise dataset show that our approach is both efficient and effective to perform indexing and retrieval tasks of motion data.

[1] K. Adistambha, S. Davis, C. Ritz, I. S. Burnett and D. Stirling, “Enhancing Multimedia Search Using Human Motion,” Multimedia - A Multidisciplinary Approach to Complex Issues, InTech, pp. 161-174, March 2012.
[2] J. Bernard, N. Wilhelm, B. Kruger, T. May, T. Schreck, and J. Kohlhammer, “MotionExplorer: Exploratory Search in Human Motion Capture Data Based on Hierarchical Aggregation,” IEEE Transactions on Visualization and Computer Graphics, 19(12):2257-2266, September 2013.
[3] S. Celebi, A. S. Aydin, T. T. Temiz, and T. Arici, “Gesture Recognition Using Skeleton Data with Weighted Dynamic Time Warping,” Proceedings of the 8th International Conference on Computer Vision Theory and Applications, February 2013.
[4] W. Choensawat, W. Choi, and K. Hachimura, “Similarity Retrieval of Motion Capture Data Based on Derivative Features,” Journal of Advanced Computational Intelligence and Intelligent Informatics, 16(1):13-23, 2012.
[5] T. Cloete and C. Scheffer, “Repeatability of an Off-the-shelf, Full Body Inertial Motion Capture System during Clinical Gait Analysis,” Proceedings of the 32nd Annual International Conference on IEEE Engineering in Medicine and Biology Society, pp. 5125-5128, August 2010.
[6] L. Deng, H. Leung, N. Gu, and Y. Yang, “Generalized Model‐Based Human Motion Recognition with Body Partition Index Maps,” Computer Graphics Forum, 31(1):202-215, February 2012.
[7] X. Guo and Q. Zhang, “3D Human Motion Retrieval Based on Human Hierarchical Index Structure,” Biology of Sport, 30(2):145, March 2013.
[8] X. Guo, Q. Zhang, R. Liu, D. Zhou, and J. Dong, “3D Human Motion Retrieval Based on ISOMAP Dimension Reduction,” Proceedings of the 3rd International Conference on Artificial Intelligence and Computational Intelligence, pp. 159-169, September 2011.
[9] S. Hamano, W. Takano, and Y. Nakamura, “Motion Data Retrieval Based on Statistic Correlation between Motion Symbol Space and Language,” Proceedings of the 2011 IEEE/RSJ International Conference on Intelligent Robots and Systems, pp. 3401-3406, September 2011.
[10] Y. Hu, S. Wu, S. Xia, J. Fu, and W. Chen, “Motion Track: Visualizing Variations of Human Motion Data,” Proceedings of the 3rd IEEE Pacific Visualization Symposium, pp. 153–160, March 2010.
[11] T. Huang, H. Liu, and G. Ding, “Motion Retrieval Based on Kinetic Features in Large Motion Database,” Proceedings of the 14th ACM International Conference on Multimodal Interaction, pp. 209-216, October 2012.
[12] V. John, E. Trucco, and S. Ivekovic, “Markerless Human Articulated Tracking Using Hierarchical Particle Swarm Optimisation,” Image and Vision Computing, 28(11):1530-1547, November 2010.
[13] T. Kohonen, “The Self-organizing Map,” Proceedings of IEEE, 78(9):1464-1480, September 1990.
[14] F. A. Kondori, S. Yousefi, H. Li, and S. Sonning, “3D Head Pose Estimation Using The Kinect,” Proceedings of the 2011 International Conference on Wireless Communications and Signal Processing, pp. 1-4, November 2011.
[15] C.-S. Lee and A. Elgammal, “Human Motion Synthesis by Motion Manifold Learning and Motion Primitive Segmentation,” Proceedings of the IV Conference on Articulated Motion and Deformable Objects, pp. 464-473, July 2006.
[16] M. Müller, “Dynamic Time Warping,” Information Retrieval for Music and Motion, Springer-Verlag, pp. 69–84, September 2007
[17] M. Müller and T. Röder, “Motion Templates for Automatic Classification and Retrieval of Motion Capture Data,” Proceedings of the 2006 ACM SIGGRAPH/Eurographics Symposium on Computer Animation, pp. 137-146, September 2006.
[18] M. Müller, T. Röder and M. Clausen, “Efficient Content-Based Retrieval of Motion Capture Data,” Proceedings of the 32nd International Conference on Computer Graphics and Interactive Techniques, pp. 677–685, July 2005.
[19] F. Negin, F. Özdemir, C. B. Akgül, K. A. Yüksel, and A. Erçil , “A Decision Forest Based Feature Selection Framework for Action Recognition from RGB-Depth Cameras,” Proceedings of the International Conference on Image Analysis and Recognition, pp. 648-657, June 2013.
[20] K.-L. Pang, “Discovering Unusual Behavior Patterns from Motion Data,” Master Thesis, Department of Engineering Science, National Cheng Kung University, Tainan, Taiwan, February 2014.
[21] PrimeSense: 3D Sensors and Natural Interaction Solutions, http://www.primesense.com/.
[22] C. Ren, X. Lei, and G. Zhang, “Motion Data Retrieval from Very Large Motion Databases,” Proceedings of International Conference on Virtual Reality and Visualization, pp. 70–77, November 2011.
[23] Y. Sakamoto, S. Kuriyama, and T. Kaneko, “Motion Map: Image-based Retrieval And Segmentation of Motion Data,” Proceedings of ACM SIGGRAPH/Eurographics Symposium on Computer Animation, pp. 259–266, August 2004.
[24] 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 Image,” Proceedings of the 2011 IEEE Conference on Computer Vision and Pattern Recognition, pp. 1297-1304, June 2011.
[25] W.-G. Teng, P.-L. Chang, and C.-T. Yang, “Adaptive and Efficient Colour Quantisation Based on a Growing Self-Organising Map,” IET Image Processing, 6(5):463-472, July 2012.
[26] K. Wongsuphasawat and B. Shneiderman, “Finding Comparable Temporal Categorical Records: A Similarity Measure with An Interactive Visualization, ” Proceedings of the IEEE Symposium on Visual Analytics Science and Technology, pp. 27–34, October 2009.
[27] S. Wu, Z. Wang, and S. Xia, “Indexing and Retrieval of Human Motion Data by a Hierarchical Tree,” Proceedings of the 16th ACM Symposium on Virtual Reality Software and Technology, pp.207-214, November 2009.
[28] S. Wu, S. Xia, Z. Wang, and C. Li, “Efficient Motion Data Indexing and Retrieval with Local Similarity Measure of Motion Strings,” The Visual Computer, 25(5-7):499-508, May 2009.
[29] Q. Xiao, Y. Luo, and S. Gao, “Human Motion Retrieval with Symbolic Aggregate Approximation,” Proceedings of the 24th Chinese Control and Decision Conference, pp. 3632-3636, May 2012.