Robust human posture analysis using incremental learning and recall based on degree of confidence of feature points

Atsushi Shimada, Madoka Kanouchi, Daisaku Arita, Rin-Ichiro Taniguchi

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Purpose - The purpose of this paper is to present an approach to improve the accuracy of estimating feature points of human body on a vision-based motion capture system (MCS) by using the variable-density self-organizing map (VDSOM). Design/methodology/approach - The VDSOM is a kind of self-organizing map (SOM) and has an ability to learn training samples incrementally. The authors let VDSOM learn 3D feature points of human body when the MCS succeeded in estimating them correctly. On the other hand, one or more 3D feature point could not be estimated correctly, the VDSOM is used for the other purpose. The SOM including VDSOM has an ability to recall a part of weight vector which have learned in the learning process. This ability is used to recall correct patterns and complement such incorrect feature points by replacing such incorrect feature points with them. Findings - Experimental results show that the approach is effective for estimation of human posture robustly compared with the other methods. Originality/value - The proposed approach is interesting for the collaboration between an MCS and an incremental learning.

Original languageEnglish
Pages (from-to)304-326
Number of pages23
JournalInternational Journal of Intelligent Computing and Cybernetics
Volume2
Issue number2
DOIs
Publication statusPublished - Jun 5 2009

Fingerprint

Self organizing maps

All Science Journal Classification (ASJC) codes

  • Computer Science(all)

Cite this

@article{87891dabc02b4946b1116be7c7a8b9af,
title = "Robust human posture analysis using incremental learning and recall based on degree of confidence of feature points",
abstract = "Purpose - The purpose of this paper is to present an approach to improve the accuracy of estimating feature points of human body on a vision-based motion capture system (MCS) by using the variable-density self-organizing map (VDSOM). Design/methodology/approach - The VDSOM is a kind of self-organizing map (SOM) and has an ability to learn training samples incrementally. The authors let VDSOM learn 3D feature points of human body when the MCS succeeded in estimating them correctly. On the other hand, one or more 3D feature point could not be estimated correctly, the VDSOM is used for the other purpose. The SOM including VDSOM has an ability to recall a part of weight vector which have learned in the learning process. This ability is used to recall correct patterns and complement such incorrect feature points by replacing such incorrect feature points with them. Findings - Experimental results show that the approach is effective for estimation of human posture robustly compared with the other methods. Originality/value - The proposed approach is interesting for the collaboration between an MCS and an incremental learning.",
author = "Atsushi Shimada and Madoka Kanouchi and Daisaku Arita and Rin-Ichiro Taniguchi",
year = "2009",
month = "6",
day = "5",
doi = "10.1108/17563780910959910",
language = "English",
volume = "2",
pages = "304--326",
journal = "International Journal of Intelligent Computing and Cybernetics",
issn = "1756-378X",
publisher = "Emerald Group Publishing Ltd.",
number = "2",

}

TY - JOUR

T1 - Robust human posture analysis using incremental learning and recall based on degree of confidence of feature points

AU - Shimada, Atsushi

AU - Kanouchi, Madoka

AU - Arita, Daisaku

AU - Taniguchi, Rin-Ichiro

PY - 2009/6/5

Y1 - 2009/6/5

N2 - Purpose - The purpose of this paper is to present an approach to improve the accuracy of estimating feature points of human body on a vision-based motion capture system (MCS) by using the variable-density self-organizing map (VDSOM). Design/methodology/approach - The VDSOM is a kind of self-organizing map (SOM) and has an ability to learn training samples incrementally. The authors let VDSOM learn 3D feature points of human body when the MCS succeeded in estimating them correctly. On the other hand, one or more 3D feature point could not be estimated correctly, the VDSOM is used for the other purpose. The SOM including VDSOM has an ability to recall a part of weight vector which have learned in the learning process. This ability is used to recall correct patterns and complement such incorrect feature points by replacing such incorrect feature points with them. Findings - Experimental results show that the approach is effective for estimation of human posture robustly compared with the other methods. Originality/value - The proposed approach is interesting for the collaboration between an MCS and an incremental learning.

AB - Purpose - The purpose of this paper is to present an approach to improve the accuracy of estimating feature points of human body on a vision-based motion capture system (MCS) by using the variable-density self-organizing map (VDSOM). Design/methodology/approach - The VDSOM is a kind of self-organizing map (SOM) and has an ability to learn training samples incrementally. The authors let VDSOM learn 3D feature points of human body when the MCS succeeded in estimating them correctly. On the other hand, one or more 3D feature point could not be estimated correctly, the VDSOM is used for the other purpose. The SOM including VDSOM has an ability to recall a part of weight vector which have learned in the learning process. This ability is used to recall correct patterns and complement such incorrect feature points by replacing such incorrect feature points with them. Findings - Experimental results show that the approach is effective for estimation of human posture robustly compared with the other methods. Originality/value - The proposed approach is interesting for the collaboration between an MCS and an incremental learning.

UR - http://www.scopus.com/inward/record.url?scp=78549256477&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=78549256477&partnerID=8YFLogxK

U2 - 10.1108/17563780910959910

DO - 10.1108/17563780910959910

M3 - Article

AN - SCOPUS:78549256477

VL - 2

SP - 304

EP - 326

JO - International Journal of Intelligent Computing and Cybernetics

JF - International Journal of Intelligent Computing and Cybernetics

SN - 1756-378X

IS - 2

ER -