Extraction and Description of Moving Human Body by Periodic Motion Analysis


Yoo, Jang-Hee, Nixon, Mark S. and Harris, Chris J. (2002) Extraction and Description of Moving Human Body by Periodic Motion Analysis. At Proc. ISCA 17th International Conference on Computers and Their Applications The International Society for Computers and Their Applications-ISCA, 110-113.

Download

[img] Postscript
Download (1048Kb)

Description/Abstract

We describe a new method for extraction and description of a moving human body by periodic motion analysis from image sequences. The moving human body is detected by background subtraction using an edge difference image with noise filtering. Then, histogram projection profiles of the edge difference image are used to determine the position of the human body. This method is improved by periodic motion analysis using projection profiles in a moving window. The periodicity of human gait motion is described by a gait state diagram and primitive during a gait cycle. The usefulness of proposed method is demonstrated in experiments with comparison of other methods.

Item Type: Conference or Workshop Item (Speech)
Additional Information: Organisation: ISCA
ISBNs: 1880843420
Divisions: Faculty of Physical and Applied Science > Electronics and Computer Science > Comms, Signal Processing & Control
Item ID: 256501
Date Deposited: 20 Nov 2003
Last Modified: 01 Mar 2012 10:48
Contributors: Yoo, Jang-Hee (Author)
Nixon, Mark S. (Author)
Harris, Chris J. (Author)
Date: April 2002
Additional Information: Organisation: ISCA
Status: Published
Publisher: The International Society for Computers and Their Applications-ISCA
Further Information:Google Scholar
URI: http://eprints.soton.ac.uk/id/eprint/256501

Actions (login required)

View Item View Item