Human Activity Classification Based on Gait Energy
Image and Coevolutionary Genetic Programming
X. Zou, and B. Bhanu. 18th International Conference on Pattern Recognition
(ICPR'06), III, page 556--559. Hong Kong, IEEE, (20-24 August 2006)
DOI: doi:10.1109/ICPR.2006.633
Abstract
In this paper, we present a novel approach based on
gait energy image (GEI) and co-evolutionary genetic
programming (CGP) for human activity classification.
Specifically, Hu's moment and normalized histogram bins
are extracted from the original GEIs as input features.
CGP is employed to reduce the feature dimensionality
and learn the classifiers. The strategy of majority
voting is applied to the CGP to improve the overall
performance in consideration of the diversification of
genetic programming. This learningbased approach
improves the classification accuracy by approximately 7
percent in comparison to the traditional classifiers.
%0 Conference Paper
%1 bb51319
%A Zou, Xiaotao
%A Bhanu, Bir
%B 18th International Conference on Pattern Recognition
(ICPR'06)
%C Hong Kong
%D 2006
%E Tang, Yuan Yan
%E Wang, Patrick
%E Lorette, G.
%E Yeung, Daniel So
%I IEEE
%K algorithms, coevolution genetic programming,
%P 556--559
%R doi:10.1109/ICPR.2006.633
%T Human Activity Classification Based on Gait Energy
Image and Coevolutionary Genetic Programming
%U http://doi.ieeecomputersociety.org/10.1109/ICPR.2006.633
%V III
%X In this paper, we present a novel approach based on
gait energy image (GEI) and co-evolutionary genetic
programming (CGP) for human activity classification.
Specifically, Hu's moment and normalized histogram bins
are extracted from the original GEIs as input features.
CGP is employed to reduce the feature dimensionality
and learn the classifiers. The strategy of majority
voting is applied to the CGP to improve the overall
performance in consideration of the diversification of
genetic programming. This learningbased approach
improves the classification accuracy by approximately 7
percent in comparison to the traditional classifiers.
@inproceedings{bb51319,
abstract = {In this paper, we present a novel approach based on
gait energy image (GEI) and co-evolutionary genetic
programming (CGP) for human activity classification.
Specifically, Hu's moment and normalized histogram bins
are extracted from the original GEIs as input features.
CGP is employed to reduce the feature dimensionality
and learn the classifiers. The strategy of majority
voting is applied to the CGP to improve the overall
performance in consideration of the diversification of
genetic programming. This learningbased approach
improves the classification accuracy by approximately 7
percent in comparison to the traditional classifiers.},
added-at = {2008-06-19T17:46:40.000+0200},
address = {Hong Kong},
author = {Zou, Xiaotao and Bhanu, Bir},
bibsource = {http://iris.usc.edu/Vision-Notes/bibliography/motion-f730.html#TT48462},
biburl = {https://www.bibsonomy.org/bibtex/27ae39c2642b6d47e79cb73ce0b0044b6/brazovayeye},
booktitle = {18th International Conference on Pattern Recognition
(ICPR'06)},
doi = {doi:10.1109/ICPR.2006.633},
editor = {Tang, Yuan Yan and Wang, Patrick and Lorette, G. and Yeung, Daniel So},
interhash = {530e5d5084313c9765f63230b6827a56},
intrahash = {7ae39c2642b6d47e79cb73ce0b0044b6},
keywords = {algorithms, coevolution genetic programming,},
month = {20-24 August},
notes = {author is Xiaoli Zhou?
http://www.comp.hkbu.edu.hk/~icpr06/accepted.php?track=2},
organisation = {ICPR},
pages = {556--559},
publisher = {IEEE},
timestamp = {2008-06-19T17:55:53.000+0200},
title = {Human Activity Classification Based on Gait Energy
Image and Coevolutionary Genetic Programming},
url = {http://doi.ieeecomputersociety.org/10.1109/ICPR.2006.633},
volume = {III},
year = 2006
}