Semantically Consistent Human Motion Segmentation

The development of motion capturing devices like Microsoft Kinect poses new challenges in the exploitation of human-motion data for various application fields, such as computer animation, visual surveillance, sports or physical medicine. In such applications, motion segmentation is recognized as one of the most fundamental steps. Existing methods usually segment motions at the level of logical actions, like walking or jumping, to annotate the motion segments by textual descriptions. Although the action-level segmentation is convenient for motion summarization and action retrieval, it does not suit for general action-independent motion retrieval. In this paper, we introduce a novel semantically consistent algorithm for partitioning motions into short and further non-divisible segments. The property of semantic consistency ensures that the start and end of each segment are detected at semantically equivalent phases of movement to support general motion retrieval. The proposed segmentation algorithm first extracts relative distances between particular body parts as motion features. Based on these features, segments are consequently identified by constructing and analyzing a one-dimensional energy curve representing local motion changes. Experiments conducted on real-life motions demonstrate that the algorithm outperforms other relevant approaches in terms of recall and precision with respect to a user-defined ground truth. Moreover, it identifies segments at semantically equivalent phases with the highest accuracy.

[1]  Adam Świtoński,et al.  Motion capture as Data Source for Gait-based Human Identification , 2012 .

[2]  Adso Fernández-Baena,et al.  Biomechanical Validation of Upper-Body and Lower-Body Joint Movements of Kinect Motion Capture Data for Rehabilitation Treatments , 2012, 2012 Fourth International Conference on Intelligent Networking and Collaborative Systems.

[3]  Pavel Zezula,et al.  A Key-Pose Similarity Algorithm for Motion Data Retrieval , 2013, ACIVS.

[4]  F. Xavier Roca,et al.  Automatic Key Pose Selection for 3D Human Action Recognition , 2010, AMDO.

[5]  Tido Röder,et al.  Documentation Mocap Database HDM05 , 2007 .

[6]  Jehee Lee,et al.  Finding Syntactic Structures from Human Motion Data , 2011, Comput. Graph. Forum.

[7]  Xiaoyong Lei,et al.  Motion Data Retrieval from Very Large Motion Databases , 2011, 2011 International Conference on Virtual Reality and Visualization.

[8]  Jessica K. Hodgins,et al.  Hierarchical Aligned Cluster Analysis for Temporal Clustering of Human Motion , 2013, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[9]  Yueting Zhuang,et al.  An Efficient Keyframe Extraction from Motion Capture Data , 2006, Computer Graphics International.

[10]  Hans-Peter Seidel,et al.  Efficient and Robust Annotation of Motion Capture Data , 2009 .

[11]  Yi Lin,et al.  Nonuniform Segment-Based Compression of Motion Capture Data , 2007, ISVC.

[12]  B. Prabhakaran,et al.  Segmentation and recognition of motion capture data stream by classification , 2007, Multimedia Tools and Applications.

[13]  Bodo Rosenhahn,et al.  An Evolutionary Approach for Learning Motion Class Patterns , 2008, DAGM-Symposium.

[14]  Jinsheng Kang,et al.  Human motion modeling and simulation , 2006 .

[15]  Woong Choi,et al.  Similarity Retrieval of Motion Capture Data Based on Derivative Features , 2012, Journal of Advanced Computational Intelligence and Intelligent Informatics.

[16]  Dimitrios Gunopulos,et al.  Indexing Large Human-Motion Databases , 2004, VLDB.

[17]  Ramakant Nevatia,et al.  Recognition and Segmentation of 3-D Human Action Using HMM and Multi-class AdaBoost , 2006, ECCV.

[18]  Pavel Zezula,et al.  Gait Recognition Based on Normalized Walk Cycles , 2012, ISVC.

[19]  Feng Liu,et al.  3D motion retrieval with motion index tree , 2003, Comput. Vis. Image Underst..

[20]  Huaijiang Sun,et al.  Text-Like Motion Representation for Human Motion Retrieval , 2012, IScIDE.

[21]  Eamonn J. Keogh,et al.  Scaling and time warping in time series querying , 2005, The VLDB Journal.

[22]  Changyin Sun,et al.  Intelligent Science and Intelligent Data Engineering , 2011, Lecture Notes in Computer Science.

[23]  Antti Oulasvirta,et al.  Computer Vision – ECCV 2006 , 2006, Lecture Notes in Computer Science.

[24]  Zhaoqi Wang,et al.  Indexing and retrieval of human motion data by a hierarchical tree , 2009, VRST '09.

[25]  Okan Arikan,et al.  Interactive motion generation from examples , 2002, ACM Trans. Graph..

[26]  Jernej Barbic,et al.  Segmenting Motion Capture Data into Distinct Behaviors , 2004, Graphics Interface.

[27]  Rynson W. H. Lau,et al.  A real-time database architecture for motion capture data , 2011, ACM Multimedia.

[28]  D. Cohen-Or,et al.  Action synopsis: pose selection and illustration , 2005, SIGGRAPH 2005.

[29]  Huaijiang Sun,et al.  Automated human motion segmentation via motion regularities , 2013, The Visual Computer.

[30]  Hans-Peter Seidel,et al.  Advances in Computer Graphics, 24th Computer Graphics International Conference, CGI 2006, Hangzhou, China, June 26-28, 2006, Proceedings , 2006, CGI.