Robust design of face recognition systems

Sunjin Yu, Hyobin Lee, Jaihie Kim, Sangyoun Lee

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Currently, most face recognition methods provide a number of parameters to be optimized, leaving the selection and optimization of the right parameter set is necessary for the implementation. The choice of the right parameter set that is suitable for a rich enough class of input faces in pose and illumination variations is, however, quite difficult. We propose robust parameter estimation, using the Taguchi method, when applied to 2nd order mixture of eigenfaces method that allows effective (near optimal) performance under pose and illumination variations. A number of experimental results confirm the improvement (via robustness) vis-'a-vis conventional parameter estimation methods, and these methods promise a solution to the design of efficient parameter sets that support many multi-variable face recognition systems.

Original languageEnglish
Title of host publicationComputational Science and Its Applications - ICCSA 2006
Subtitle of host publicationInternational Conference, Proceedings - Part II
PublisherSpringer Verlag
Pages96-105
Number of pages10
ISBN (Print)3540340726, 9783540340720
Publication statusPublished - 2006 Jan 1
EventICCSA 2006: International Conference on Computational Science and Its Applications - Glasgow, United Kingdom
Duration: 2006 May 82006 May 11

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume3981 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

OtherICCSA 2006: International Conference on Computational Science and Its Applications
CountryUnited Kingdom
CityGlasgow
Period06/5/806/5/11

Fingerprint

Robust Design
Face recognition
Face Recognition
Parameter estimation
Lighting
Taguchi methods
Robustness (control systems)
Parameter Estimation
Illumination
Eigenface
Taguchi Method
Robust Estimation
Robustness
Necessary
Optimization
Experimental Results

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Yu, S., Lee, H., Kim, J., & Lee, S. (2006). Robust design of face recognition systems. In Computational Science and Its Applications - ICCSA 2006: International Conference, Proceedings - Part II (pp. 96-105). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 3981 LNCS). Springer Verlag.
Yu, Sunjin ; Lee, Hyobin ; Kim, Jaihie ; Lee, Sangyoun. / Robust design of face recognition systems. Computational Science and Its Applications - ICCSA 2006: International Conference, Proceedings - Part II. Springer Verlag, 2006. pp. 96-105 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{5ca0a10d6b3c4d7591aa4a979a54e6ac,
title = "Robust design of face recognition systems",
abstract = "Currently, most face recognition methods provide a number of parameters to be optimized, leaving the selection and optimization of the right parameter set is necessary for the implementation. The choice of the right parameter set that is suitable for a rich enough class of input faces in pose and illumination variations is, however, quite difficult. We propose robust parameter estimation, using the Taguchi method, when applied to 2nd order mixture of eigenfaces method that allows effective (near optimal) performance under pose and illumination variations. A number of experimental results confirm the improvement (via robustness) vis-'a-vis conventional parameter estimation methods, and these methods promise a solution to the design of efficient parameter sets that support many multi-variable face recognition systems.",
author = "Sunjin Yu and Hyobin Lee and Jaihie Kim and Sangyoun Lee",
year = "2006",
month = "1",
day = "1",
language = "English",
isbn = "3540340726",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "96--105",
booktitle = "Computational Science and Its Applications - ICCSA 2006",
address = "Germany",

}

Yu, S, Lee, H, Kim, J & Lee, S 2006, Robust design of face recognition systems. in Computational Science and Its Applications - ICCSA 2006: International Conference, Proceedings - Part II. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 3981 LNCS, Springer Verlag, pp. 96-105, ICCSA 2006: International Conference on Computational Science and Its Applications, Glasgow, United Kingdom, 06/5/8.

Robust design of face recognition systems. / Yu, Sunjin; Lee, Hyobin; Kim, Jaihie; Lee, Sangyoun.

Computational Science and Its Applications - ICCSA 2006: International Conference, Proceedings - Part II. Springer Verlag, 2006. p. 96-105 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 3981 LNCS).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Robust design of face recognition systems

AU - Yu, Sunjin

AU - Lee, Hyobin

AU - Kim, Jaihie

AU - Lee, Sangyoun

PY - 2006/1/1

Y1 - 2006/1/1

N2 - Currently, most face recognition methods provide a number of parameters to be optimized, leaving the selection and optimization of the right parameter set is necessary for the implementation. The choice of the right parameter set that is suitable for a rich enough class of input faces in pose and illumination variations is, however, quite difficult. We propose robust parameter estimation, using the Taguchi method, when applied to 2nd order mixture of eigenfaces method that allows effective (near optimal) performance under pose and illumination variations. A number of experimental results confirm the improvement (via robustness) vis-'a-vis conventional parameter estimation methods, and these methods promise a solution to the design of efficient parameter sets that support many multi-variable face recognition systems.

AB - Currently, most face recognition methods provide a number of parameters to be optimized, leaving the selection and optimization of the right parameter set is necessary for the implementation. The choice of the right parameter set that is suitable for a rich enough class of input faces in pose and illumination variations is, however, quite difficult. We propose robust parameter estimation, using the Taguchi method, when applied to 2nd order mixture of eigenfaces method that allows effective (near optimal) performance under pose and illumination variations. A number of experimental results confirm the improvement (via robustness) vis-'a-vis conventional parameter estimation methods, and these methods promise a solution to the design of efficient parameter sets that support many multi-variable face recognition systems.

UR - http://www.scopus.com/inward/record.url?scp=33745929333&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33745929333&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:33745929333

SN - 3540340726

SN - 9783540340720

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 96

EP - 105

BT - Computational Science and Its Applications - ICCSA 2006

PB - Springer Verlag

ER -

Yu S, Lee H, Kim J, Lee S. Robust design of face recognition systems. In Computational Science and Its Applications - ICCSA 2006: International Conference, Proceedings - Part II. Springer Verlag. 2006. p. 96-105. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).