Learning Robust Features for Gait Recognition by Maximum Margin
Criterion

D 2016

Learning Robust Features for Gait Recognition by Maximum Margin Criterion

BALÁŽIA, Michal and Petr SOJKA

Basic information

Original name

Learning Robust Features for Gait Recognition by Maximum Margin Criterion

Authors

BALÁŽIA, Michal (703 Slovakia, guarantor, belonging to the institution) and Petr SOJKA (203 Czech Republic, belonging to the institution)

Edition

USA, Proceedings of the 23rd IEEE/IAPR International Conference on Pattern Recognition (ICPR 2016), p. 901-906, 6 pp. 2016

Publisher

IEEE

Other information

Language

English

Type of outcome

Stať ve sborníku

Field of Study

10201 Computer sciences, information science, bioinformatics

Country of publisher

United States of America

Confidentiality degree

není předmětem státního či obchodního tajemství

Publication form

electronic version available online

References:

DOI conference web Proceedings

RIV identification code

RIV/00216224:14330/16:00090367

Organization unit

Faculty of Informatics

ISBN

978-1-5090-4847-2

DOI

http://dx.doi.org/10.1109/ICPR.2016.7899750

UT WoS

000406771300153

Keywords (in Czech)

rozpoznávání podle chůze

Keywords in English

gait recognition

Abstract

V originále

In the field of gait recognition from motion capture data, designing human-interpretable gait features is a common practice of many fellow researchers. To refrain from ad-hoc schemes and to find maximally discriminative features we may need to explore beyond the limits of human interpretability. This paper contributes to the state-of-the-art with a machine learning approach for extracting robust gait features directly from raw joint coordinates. The features are learned by a modification of Linear Discriminant Analysis with Maximum Margin Criterion so that the identities are maximally separated and, in combination with an appropriate classifier, used for gait recognition. Experiments on the CMU MoCap database show that this method outperforms eight other relevant methods in terms of the distribution of biometric templates in respective feature spaces expressed in four class separability coefficients. Additional experiments indicate that this method is a leading concept for rank-based classifier systems.

Links

MUNI/A/0892/2015, interní kód MU

Name: Výzkum v aplikované informatice na FI MU (Acronym: VAIFIMU)

Investor: Masaryk University, Category A

MUNI/A/0935/2015, interní kód MU

Name: Zapojení studentů Fakulty informatiky do mezinárodní vědecké komunity (Acronym: SKOMU)

Investor: Masaryk University, Category A

Files attached

icpr679arxiv.pdf

Find similar documents

Citovat

BALÁŽIA, Michal and Petr SOJKA. Learning Robust Features for Gait Recognition by Maximum Margin Criterion. Online. In Eduardo Bayro-Corrochano, Gabrielle Sanniti di Baja, Gérard Medioni. Proceedings of the 23rd IEEE/IAPR International Conference on Pattern Recognition (ICPR 2016). USA: IEEE, 2016, p. 901-906. ISBN 978-1-5090-4847-2. Available from: https://dx.doi.org/10.1109/ICPR.2016.7899750.

@inproceedings{1349591,
   author = {Balážia, Michal and Sojka, Petr},
   address = {USA},
   booktitle = {Proceedings of the 23rd IEEE/IAPR International Conference on Pattern Recognition (ICPR 2016)},
   doi = {http://dx.doi.org/10.1109/ICPR.2016.7899750},
   editor = {Eduardo Bayro-Corrochano, Gabrielle Sanniti di Baja, Gérard Medioni},
   keywords = {gait recognition},
   howpublished = {elektronická verze "online"},
   language = {eng},
   location = {USA},
   isbn = {978-1-5090-4847-2},
   pages = {901-906},
   publisher = {IEEE},
   title = {Learning Robust Features for Gait Recognition by Maximum Margin Criterion},
   url = {https://doi.org/10.1109/ICPR.2016.7899750},
   year = {2016}
}

TY  - JOUR
ID  - 1349591
AU  - Balážia, Michal - Sojka, Petr
PY  - 2016
TI  - Learning Robust Features for Gait Recognition by Maximum Margin Criterion
PB  - IEEE
CY  - USA
SN  - 9781509048472
KW  - gait recognition
UR  - https://doi.org/10.1109/ICPR.2016.7899750
L2  - https://doi.org/10.1109/ICPR.2016.7899750
N2  - In the field of gait recognition from motion capture data, designing human-interpretable gait features is a common practice of many fellow researchers. To refrain from ad-hoc schemes and to find maximally discriminative features we may need to explore beyond the limits of human interpretability. This paper contributes to the state-of-the-art with a machine learning approach for extracting robust gait features directly from raw joint coordinates. The features are learned by a modification of Linear Discriminant Analysis with Maximum Margin Criterion so that the identities are maximally separated and, in combination with an appropriate classifier, used for gait recognition. Experiments on the CMU MoCap database show that this method outperforms eight other relevant methods in terms of the distribution of biometric templates in respective feature spaces expressed in four class separability coefficients. Additional experiments indicate that this method is a leading concept for rank-based classifier systems.
ER  -

BALÁŽIA, Michal and Petr SOJKA. Learning Robust Features for Gait Recognition by Maximum Margin Criterion. Online. In Eduardo Bayro-Corrochano, Gabrielle Sanniti di Baja, Gérard Medioni. \textit{Proceedings of the 23rd IEEE/IAPR International Conference on Pattern Recognition (ICPR 2016)}. USA: IEEE, 2016, p.~901-906. ISBN~978-1-5090-4847-2. Available from: https://dx.doi.org/10.1109/ICPR.2016.7899750.

Detailed Information on Publication Record