Learning-based image representation and method for face recognition

Zhiming Liu; Chengjun Liu; Qingchuan Tao

doi:10.1109/BTAS.2009.5339012

Learning-based image representation and method for face recognition

Zhiming Liu, Chengjun Liu, Qingchuan Tao

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

2 Scopus citations

Abstract

This paper presents a novel method for face recognition. First, we generate the new image representation from the decorrelated hybrid color configurations rather than RGB color space via a learning algorithm. The learning algorithm, Principal Component Analysis (PCA) plus Fisher Linear Discriminant analysis (FLD), is able to derive the desired color transformation to generate a discriminating image representation that is optimal for face recognition. Second, we partition face image into some small patches, each of which can obtain its own color transformation, to reduce the effect of illumination variations. Thus, a patch-based novel image representation method is proposed for face recognition. Experiments on the Face Recognition Grand Challenge (FRGC) version 2 Experiment 4 show that the proposed method outperforms gray-scale image and some recent methods in face recognition.

Original language	English (US)
Title of host publication	IEEE 3rd International Conference on Biometrics
Subtitle of host publication	Theory, Applications and Systems, BTAS 2009
DOIs	https://doi.org/10.1109/BTAS.2009.5339012
State	Published - 2009
Externally published	Yes
Event	IEEE 3rd International Conference on Biometrics: Theory, Applications and Systems, BTAS 2009 - Washington, DC, United States Duration: Sep 28 2009 → Sep 30 2009

Publication series

Name	IEEE 3rd International Conference on Biometrics: Theory, Applications and Systems, BTAS 2009

Other

Other	IEEE 3rd International Conference on Biometrics: Theory, Applications and Systems, BTAS 2009
Country/Territory	United States
City	Washington, DC
Period	9/28/09 → 9/30/09

All Science Journal Classification (ASJC) codes

Biotechnology
Computational Theory and Mathematics
Computer Vision and Pattern Recognition

Access to Document

10.1109/BTAS.2009.5339012

Cite this

@inproceedings{6240ea61ad2049c2b98ce09251f8d20c,

title = "Learning-based image representation and method for face recognition",

abstract = "This paper presents a novel method for face recognition. First, we generate the new image representation from the decorrelated hybrid color configurations rather than RGB color space via a learning algorithm. The learning algorithm, Principal Component Analysis (PCA) plus Fisher Linear Discriminant analysis (FLD), is able to derive the desired color transformation to generate a discriminating image representation that is optimal for face recognition. Second, we partition face image into some small patches, each of which can obtain its own color transformation, to reduce the effect of illumination variations. Thus, a patch-based novel image representation method is proposed for face recognition. Experiments on the Face Recognition Grand Challenge (FRGC) version 2 Experiment 4 show that the proposed method outperforms gray-scale image and some recent methods in face recognition.",

author = "Zhiming Liu and Chengjun Liu and Qingchuan Tao",

year = "2009",

doi = "10.1109/BTAS.2009.5339012",

language = "English (US)",

isbn = "9781424450206",

series = "IEEE 3rd International Conference on Biometrics: Theory, Applications and Systems, BTAS 2009",

booktitle = "IEEE 3rd International Conference on Biometrics",

note = "IEEE 3rd International Conference on Biometrics: Theory, Applications and Systems, BTAS 2009 ; Conference date: 28-09-2009 Through 30-09-2009",

}

Liu, Z, Liu, C & Tao, Q 2009, Learning-based image representation and method for face recognition. in IEEE 3rd International Conference on Biometrics: Theory, Applications and Systems, BTAS 2009., 5339012, IEEE 3rd International Conference on Biometrics: Theory, Applications and Systems, BTAS 2009, IEEE 3rd International Conference on Biometrics: Theory, Applications and Systems, BTAS 2009, Washington, DC, United States, 9/28/09. https://doi.org/10.1109/BTAS.2009.5339012

Learning-based image representation and method for face recognition. / Liu, Zhiming; Liu, Chengjun; Tao, Qingchuan.
IEEE 3rd International Conference on Biometrics: Theory, Applications and Systems, BTAS 2009. 2009. 5339012 (IEEE 3rd International Conference on Biometrics: Theory, Applications and Systems, BTAS 2009).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Learning-based image representation and method for face recognition

AU - Liu, Zhiming

AU - Liu, Chengjun

AU - Tao, Qingchuan

PY - 2009

Y1 - 2009

N2 - This paper presents a novel method for face recognition. First, we generate the new image representation from the decorrelated hybrid color configurations rather than RGB color space via a learning algorithm. The learning algorithm, Principal Component Analysis (PCA) plus Fisher Linear Discriminant analysis (FLD), is able to derive the desired color transformation to generate a discriminating image representation that is optimal for face recognition. Second, we partition face image into some small patches, each of which can obtain its own color transformation, to reduce the effect of illumination variations. Thus, a patch-based novel image representation method is proposed for face recognition. Experiments on the Face Recognition Grand Challenge (FRGC) version 2 Experiment 4 show that the proposed method outperforms gray-scale image and some recent methods in face recognition.

AB - This paper presents a novel method for face recognition. First, we generate the new image representation from the decorrelated hybrid color configurations rather than RGB color space via a learning algorithm. The learning algorithm, Principal Component Analysis (PCA) plus Fisher Linear Discriminant analysis (FLD), is able to derive the desired color transformation to generate a discriminating image representation that is optimal for face recognition. Second, we partition face image into some small patches, each of which can obtain its own color transformation, to reduce the effect of illumination variations. Thus, a patch-based novel image representation method is proposed for face recognition. Experiments on the Face Recognition Grand Challenge (FRGC) version 2 Experiment 4 show that the proposed method outperforms gray-scale image and some recent methods in face recognition.

UR - http://www.scopus.com/inward/record.url?scp=71749117522&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=71749117522&partnerID=8YFLogxK

U2 - 10.1109/BTAS.2009.5339012

DO - 10.1109/BTAS.2009.5339012

M3 - Conference contribution

AN - SCOPUS:71749117522

SN - 9781424450206

T3 - IEEE 3rd International Conference on Biometrics: Theory, Applications and Systems, BTAS 2009

BT - IEEE 3rd International Conference on Biometrics

T2 - IEEE 3rd International Conference on Biometrics: Theory, Applications and Systems, BTAS 2009

Y2 - 28 September 2009 through 30 September 2009

ER -

Learning-based image representation and method for face recognition

Abstract

Publication series

Other

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this