Structure modelling of the human body using FGMM

Human motion analysis and posture recognition have received increasing attention in the area of computer vision, for example, human-machine interaction. Structure modelling and analysis of human body is important for human motion analysis and posture recognition. In this paper, we propose a novel method based on fuzzy Gaussian mixture models (FGMM) to handle the task of structure modelling. To be more specific, Microsoft Kinect Sensor firstly captures raw 3D data of human body, followed by some data preprocessing steps, which include separating human body from the background and projection from 3D to 2D, etc. Then FGMM is applied to approximate the distribution of the processed 2D points. Experiments have been conducted to verify the effectiveness of proposed method on body structure modelling.

[1]  Wen-June Wang,et al.  Human Posture Recognition Based on Images Captured by the Kinect Sensor , 2016 .

[2]  Changshui Zhang,et al.  Competitive EM algorithm for finite mixture models , 2004, Pattern Recognit..

[3]  Honghai Liu,et al.  Fuzzy Gaussian Mixture Models , 2012, Pattern Recognit..

[4]  Hassen Mekki,et al.  KUKA robot control based Kinect image analysis , 2013, 2013 International Conference on Individual and Collective Behaviors in Robotics (ICBR).

[5]  J. MacQueen Some methods for classification and analysis of multivariate observations , 1967 .

[6]  Changshui Zhang,et al.  Active curve axis Gaussian mixture models , 2005, Pattern Recognit..

[7]  Mengyin Fu,et al.  Robot manipulator self-identification for surrounding obstacle detection , 2016, Multimedia Tools and Applications.

[8]  Isaac Cohen,et al.  Posture and Gesture Recognition using 3D Body Shapes Decomposition , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) - Workshops.

[9]  Monique Thonnat,et al.  Posture recognition with a 3D human model , 2005 .

[10]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[11]  Reinhard Koch,et al.  Nonlinear Body Pose Estimation from Depth Images , 2005, DAGM-Symposium.

[12]  Masanobu Yamamoto,et al.  Human motion analysis based on a robot arm model , 1991, Proceedings. 1991 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[13]  J. Bezdek,et al.  FCM: The fuzzy c-means clustering algorithm , 1984 .