Face recognition using the second-order mixture-of-eigenfaces method

Hyun Chul Kim, Daijin Kim, Sung Yang Bang, Sang Youn Lee

Research output: Contribution to journalArticle

16 Citations (Scopus)

Abstract

The well-known eigenface method uses an eigenface set obtained from principal component analysis. However, the single eigenface set is not enough to represent the complicated face images with large variations of poses and/or illuminations. To overcome this weakness, we propose a second-order mixture-of-eigenfaces method that combines the second-order eigenface method (ISO MPG m5750, Noordwijkerhout, March 2000) and the mixture-of-eigenfaces method (a.k.a. Gaussian mixture model (Proceedings IJCNN2001, 2001). In this method, we use a couple of mixtures of multiple eigenface sets: one is a mixture of multiple approximate eigenface sets for face images and another is a mixture of multiple residual eigenface sets for residual face images. Each mixture of multiple eigenface sets has been obtained from expectation maximization learning consecutively. Based on two mixture of multiple eigenface sets, each face image is represented by a couple of feature vectors obtained by projecting the face image onto a selected approximate eigenface set and then by projecting the residual face image onto a selected residual eigenface set. Recognition is performed by the distance in the feature space between the input image and the template image stored in the face database. Simulation results show that the proposed second-order mixture-of-eigenfaces method is best for face images with illumination variations and the mixture-of-eigenfaces method is best for the face images with pose variations in terms of average of the normalized modified retrieval rank and false identification rate.

Original languageEnglish
Pages (from-to)337-349
Number of pages13
JournalPattern Recognition
Volume37
Issue number2
DOIs
Publication statusPublished - 2004 Feb

Fingerprint

Face recognition
Lighting
Principal component analysis

All Science Journal Classification (ASJC) codes

  • Software
  • Signal Processing
  • Computer Vision and Pattern Recognition
  • Artificial Intelligence

Cite this

Kim, Hyun Chul ; Kim, Daijin ; Bang, Sung Yang ; Lee, Sang Youn. / Face recognition using the second-order mixture-of-eigenfaces method. In: Pattern Recognition. 2004 ; Vol. 37, No. 2. pp. 337-349.
@article{5f6a8885c6a648ad9ed16d2bb5f85b94,
title = "Face recognition using the second-order mixture-of-eigenfaces method",
abstract = "The well-known eigenface method uses an eigenface set obtained from principal component analysis. However, the single eigenface set is not enough to represent the complicated face images with large variations of poses and/or illuminations. To overcome this weakness, we propose a second-order mixture-of-eigenfaces method that combines the second-order eigenface method (ISO MPG m5750, Noordwijkerhout, March 2000) and the mixture-of-eigenfaces method (a.k.a. Gaussian mixture model (Proceedings IJCNN2001, 2001). In this method, we use a couple of mixtures of multiple eigenface sets: one is a mixture of multiple approximate eigenface sets for face images and another is a mixture of multiple residual eigenface sets for residual face images. Each mixture of multiple eigenface sets has been obtained from expectation maximization learning consecutively. Based on two mixture of multiple eigenface sets, each face image is represented by a couple of feature vectors obtained by projecting the face image onto a selected approximate eigenface set and then by projecting the residual face image onto a selected residual eigenface set. Recognition is performed by the distance in the feature space between the input image and the template image stored in the face database. Simulation results show that the proposed second-order mixture-of-eigenfaces method is best for face images with illumination variations and the mixture-of-eigenfaces method is best for the face images with pose variations in terms of average of the normalized modified retrieval rank and false identification rate.",
author = "Kim, {Hyun Chul} and Daijin Kim and Bang, {Sung Yang} and Lee, {Sang Youn}",
year = "2004",
month = "2",
doi = "10.1016/S0031-3203(03)00227-9",
language = "English",
volume = "37",
pages = "337--349",
journal = "Pattern Recognition",
issn = "0031-3203",
publisher = "Elsevier Limited",
number = "2",

}

Face recognition using the second-order mixture-of-eigenfaces method. / Kim, Hyun Chul; Kim, Daijin; Bang, Sung Yang; Lee, Sang Youn.

In: Pattern Recognition, Vol. 37, No. 2, 02.2004, p. 337-349.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Face recognition using the second-order mixture-of-eigenfaces method

AU - Kim, Hyun Chul

AU - Kim, Daijin

AU - Bang, Sung Yang

AU - Lee, Sang Youn

PY - 2004/2

Y1 - 2004/2

N2 - The well-known eigenface method uses an eigenface set obtained from principal component analysis. However, the single eigenface set is not enough to represent the complicated face images with large variations of poses and/or illuminations. To overcome this weakness, we propose a second-order mixture-of-eigenfaces method that combines the second-order eigenface method (ISO MPG m5750, Noordwijkerhout, March 2000) and the mixture-of-eigenfaces method (a.k.a. Gaussian mixture model (Proceedings IJCNN2001, 2001). In this method, we use a couple of mixtures of multiple eigenface sets: one is a mixture of multiple approximate eigenface sets for face images and another is a mixture of multiple residual eigenface sets for residual face images. Each mixture of multiple eigenface sets has been obtained from expectation maximization learning consecutively. Based on two mixture of multiple eigenface sets, each face image is represented by a couple of feature vectors obtained by projecting the face image onto a selected approximate eigenface set and then by projecting the residual face image onto a selected residual eigenface set. Recognition is performed by the distance in the feature space between the input image and the template image stored in the face database. Simulation results show that the proposed second-order mixture-of-eigenfaces method is best for face images with illumination variations and the mixture-of-eigenfaces method is best for the face images with pose variations in terms of average of the normalized modified retrieval rank and false identification rate.

AB - The well-known eigenface method uses an eigenface set obtained from principal component analysis. However, the single eigenface set is not enough to represent the complicated face images with large variations of poses and/or illuminations. To overcome this weakness, we propose a second-order mixture-of-eigenfaces method that combines the second-order eigenface method (ISO MPG m5750, Noordwijkerhout, March 2000) and the mixture-of-eigenfaces method (a.k.a. Gaussian mixture model (Proceedings IJCNN2001, 2001). In this method, we use a couple of mixtures of multiple eigenface sets: one is a mixture of multiple approximate eigenface sets for face images and another is a mixture of multiple residual eigenface sets for residual face images. Each mixture of multiple eigenface sets has been obtained from expectation maximization learning consecutively. Based on two mixture of multiple eigenface sets, each face image is represented by a couple of feature vectors obtained by projecting the face image onto a selected approximate eigenface set and then by projecting the residual face image onto a selected residual eigenface set. Recognition is performed by the distance in the feature space between the input image and the template image stored in the face database. Simulation results show that the proposed second-order mixture-of-eigenfaces method is best for face images with illumination variations and the mixture-of-eigenfaces method is best for the face images with pose variations in terms of average of the normalized modified retrieval rank and false identification rate.

UR - http://www.scopus.com/inward/record.url?scp=0242456238&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0242456238&partnerID=8YFLogxK

U2 - 10.1016/S0031-3203(03)00227-9

DO - 10.1016/S0031-3203(03)00227-9

M3 - Article

AN - SCOPUS:0242456238

VL - 37

SP - 337

EP - 349

JO - Pattern Recognition

JF - Pattern Recognition

SN - 0031-3203

IS - 2

ER -