Model Based Approach to Non-Intrusive Human Motion Capture

Modelling and Animation of human movement with the use of motion capture is a valuable process for creating true 'life-like' motion for areas such as sports science, entertainment, and the medical industry. Traditional methods involved using intrusive techniques for the purpose of capturing the location and subsequent movements of joints and limbs. This paper describes the work being carried out by the Vision Systems Group of Dublin City University, where non-intrusive human motion capture techniques are used. It outlines the fully kinematical model created and the system used to segment and track a moving human from video sequences, using vision processing means only.

[1]  Lucia Ballerini,et al.  Time-Varying Image Processing and Moving Object Recognition , 1997 .

[2]  Demetri Terzopoulos,et al.  Snakes: Active contour models , 2004, International Journal of Computer Vision.

[3]  Paul F. Whelan,et al.  Machine Vision Algorithms in Java , 2001 .

[4]  Larry S. Davis,et al.  W4: Real-Time Surveillance of People and Their Activities , 2000, IEEE Trans. Pattern Anal. Mach. Intell..

[5]  Alex Pentland,et al.  Pfinder: Real-Time Tracking of the Human Body , 1997, IEEE Trans. Pattern Anal. Mach. Intell..

[6]  K. P. Karmann,et al.  Moving object recognition using an adaptive background memory , 1990 .

[7]  D Thalmann,et al.  Using skeleton-based tracking to increase the reliability of optical motion capture. , 2001, Human movement science.

[8]  Neil Davey,et al.  Human shape recognition from snakes using neural networks , 1999, Proceedings Third International Conference on Computational Intelligence and Multimedia Applications. ICCIMA'99 (Cat. No.PR00300).