Independent component analysis of face images

Pong C. Yuen; J. H. Lai

doi:10.1007/3-540-45482-9_55

Independent component analysis of face images

Pong C. Yuen, J. H. Lai

Department of Computer Science

Research output: Chapter in book/report/conference proceeding › Conference proceeding › peer-review

5 Citations (Scopus)

Abstract

This paper addresses the problem of face recognition using independent component analysis. As the independent components (IC) are not orthogonal, to represent a face image using the determined ICs, the ICs have to be orthogonalized, where two methods, namely Gram-Schmit Method and Householder Transformation, are proposed. In addition, to find a better set of ICs for face recognition, an efficient IC selection algorithm is developed. Face images with different facial expressions, pose variations and small occlusions are selected to test the ICA face representation and the results are encouraging.

Original language	English
Title of host publication	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Editors	Tomaso Poggio, Seong-Whan Lee, Heinrich H. Bulthoff
Publisher	Springer Verlag
Pages	545-553
Number of pages	9
ISBN (Print)	3540675604, 9783540675600
DOIs	https://doi.org/10.1007/3-540-45482-9_55
Publication status	Published - 2000
Event	1st IEEE International Workshop on Biologically Motivated Computer Vision, BMCV 2000 - Seoul, Korea, Republic of Duration: 15 May 2000 → 17 May 2000

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	1811
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	1st IEEE International Workshop on Biologically Motivated Computer Vision, BMCV 2000
Country/Territory	Korea, Republic of
City	Seoul
Period	15/05/00 → 17/05/00

Access to Document

10.1007/3-540-45482-9_55

Cite this

Yuen, P. C., & Lai, J. H. (2000). Independent component analysis of face images. In T. Poggio, S.-W. Lee, & H. H. Bulthoff (Eds.), Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (pp. 545-553). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 1811). Springer Verlag. https://doi.org/10.1007/3-540-45482-9_55

Yuen, Pong C. ; Lai, J. H. / Independent component analysis of face images. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). editor / Tomaso Poggio ; Seong-Whan Lee ; Heinrich H. Bulthoff. Springer Verlag, 2000. pp. 545-553 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{650fbc200f854b8b83212a79ce5b1d46,

title = "Independent component analysis of face images",

abstract = "This paper addresses the problem of face recognition using independent component analysis. As the independent components (IC) are not orthogonal, to represent a face image using the determined ICs, the ICs have to be orthogonalized, where two methods, namely Gram-Schmit Method and Householder Transformation, are proposed. In addition, to find a better set of ICs for face recognition, an efficient IC selection algorithm is developed. Face images with different facial expressions, pose variations and small occlusions are selected to test the ICA face representation and the results are encouraging.",

author = "Yuen, {Pong C.} and Lai, {J. H.}",

note = "Publisher Copyright: {\textcopyright} Springer-Verlag Berlin Heidelberg 2000.; 1st IEEE International Workshop on Biologically Motivated Computer Vision, BMCV 2000 ; Conference date: 15-05-2000 Through 17-05-2000",

year = "2000",

doi = "10.1007/3-540-45482-9_55",

language = "English",

isbn = "3540675604",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Verlag",

pages = "545--553",

editor = "Tomaso Poggio and Seong-Whan Lee and Bulthoff, {Heinrich H.}",

booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

address = "Germany",

}

Yuen, PC & Lai, JH 2000, Independent component analysis of face images. in T Poggio, S-W Lee & HH Bulthoff (eds), Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 1811, Springer Verlag, pp. 545-553, 1st IEEE International Workshop on Biologically Motivated Computer Vision, BMCV 2000, Seoul, Korea, Republic of, 15/05/00. https://doi.org/10.1007/3-540-45482-9_55

Independent component analysis of face images. / Yuen, Pong C.; Lai, J. H.
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). ed. / Tomaso Poggio; Seong-Whan Lee; Heinrich H. Bulthoff. Springer Verlag, 2000. p. 545-553 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 1811).

Research output: Chapter in book/report/conference proceeding › Conference proceeding › peer-review

TY - GEN

T1 - Independent component analysis of face images

AU - Yuen, Pong C.

AU - Lai, J. H.

N1 - Publisher Copyright: © Springer-Verlag Berlin Heidelberg 2000.

PY - 2000

Y1 - 2000

N2 - This paper addresses the problem of face recognition using independent component analysis. As the independent components (IC) are not orthogonal, to represent a face image using the determined ICs, the ICs have to be orthogonalized, where two methods, namely Gram-Schmit Method and Householder Transformation, are proposed. In addition, to find a better set of ICs for face recognition, an efficient IC selection algorithm is developed. Face images with different facial expressions, pose variations and small occlusions are selected to test the ICA face representation and the results are encouraging.

AB - This paper addresses the problem of face recognition using independent component analysis. As the independent components (IC) are not orthogonal, to represent a face image using the determined ICs, the ICs have to be orthogonalized, where two methods, namely Gram-Schmit Method and Householder Transformation, are proposed. In addition, to find a better set of ICs for face recognition, an efficient IC selection algorithm is developed. Face images with different facial expressions, pose variations and small occlusions are selected to test the ICA face representation and the results are encouraging.

UR - http://www.scopus.com/inward/record.url?scp=84958527651&partnerID=8YFLogxK

U2 - 10.1007/3-540-45482-9_55

DO - 10.1007/3-540-45482-9_55

M3 - Conference proceeding

AN - SCOPUS:84958527651

SN - 3540675604

SN - 9783540675600

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

SP - 545

EP - 553

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

A2 - Poggio, Tomaso

A2 - Lee, Seong-Whan

A2 - Bulthoff, Heinrich H.

PB - Springer Verlag

T2 - 1st IEEE International Workshop on Biologically Motivated Computer Vision, BMCV 2000

Y2 - 15 May 2000 through 17 May 2000

ER -

Yuen PC, Lai JH. Independent component analysis of face images. In Poggio T, Lee SW, Bulthoff HH, editors, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Springer Verlag. 2000. p. 545-553. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/3-540-45482-9_55

Independent component analysis of face images

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this