Supervised Locally Linear Embedding in face recognition

Ying Han Pang, Beng Jin Teoh, Eng Kiong Wong, Fazly Salleh Abas

Research output: Contribution to conferencePaper

21 Citations (Scopus)

Abstract

Locally Linear Embedding (LLE), which has recently emerged as a powerful face feature descriptor, suffers from a limitation. That is class-specific information of data is lacked of during face analysis. Thus, we propose a supervised LLE technique, known as class-label Locally Linear Embedding (cLLE), to overcome the problem. cLLE is able to discover the nonlinearity of high-dimensional face data by minimizing the global reconstruction error of the set of all local neighbors in the data set. cLLE utilizes user class-specific information in neighborhoods selection and thus preserves the local neighborhoods. Since the locality preservation is correlated to the class discrimination, the proposed cLLE is expected superior to LLE in face recognition. Experimental results on three face databases: ORL, AR and Yale databases, demonstrate that the proposed technique obtains better recognition performance than PCA and LLE.

Original languageEnglish
DOIs
Publication statusPublished - 2008 Sep 12
EventIEEE- International Symposium on Biometrics and Security Technologies, ISBAST'08 - Islamabad, Pakistan
Duration: 2008 Apr 232008 Apr 24

Other

OtherIEEE- International Symposium on Biometrics and Security Technologies, ISBAST'08
CountryPakistan
CityIslamabad
Period08/4/2308/4/24

Fingerprint

Face recognition
Labels

All Science Journal Classification (ASJC) codes

  • Artificial Intelligence
  • Computational Theory and Mathematics
  • Computer Science Applications
  • Control and Systems Engineering

Cite this

Pang, Y. H., Teoh, B. J., Wong, E. K., & Abas, F. S. (2008). Supervised Locally Linear Embedding in face recognition. Paper presented at IEEE- International Symposium on Biometrics and Security Technologies, ISBAST'08, Islamabad, Pakistan. https://doi.org/10.1109/ISBAST.2008.4547646
Pang, Ying Han ; Teoh, Beng Jin ; Wong, Eng Kiong ; Abas, Fazly Salleh. / Supervised Locally Linear Embedding in face recognition. Paper presented at IEEE- International Symposium on Biometrics and Security Technologies, ISBAST'08, Islamabad, Pakistan.
@conference{0f29a35463ab44328f13175a4ef62413,
title = "Supervised Locally Linear Embedding in face recognition",
abstract = "Locally Linear Embedding (LLE), which has recently emerged as a powerful face feature descriptor, suffers from a limitation. That is class-specific information of data is lacked of during face analysis. Thus, we propose a supervised LLE technique, known as class-label Locally Linear Embedding (cLLE), to overcome the problem. cLLE is able to discover the nonlinearity of high-dimensional face data by minimizing the global reconstruction error of the set of all local neighbors in the data set. cLLE utilizes user class-specific information in neighborhoods selection and thus preserves the local neighborhoods. Since the locality preservation is correlated to the class discrimination, the proposed cLLE is expected superior to LLE in face recognition. Experimental results on three face databases: ORL, AR and Yale databases, demonstrate that the proposed technique obtains better recognition performance than PCA and LLE.",
author = "Pang, {Ying Han} and Teoh, {Beng Jin} and Wong, {Eng Kiong} and Abas, {Fazly Salleh}",
year = "2008",
month = "9",
day = "12",
doi = "10.1109/ISBAST.2008.4547646",
language = "English",
note = "IEEE- International Symposium on Biometrics and Security Technologies, ISBAST'08 ; Conference date: 23-04-2008 Through 24-04-2008",

}

Pang, YH, Teoh, BJ, Wong, EK & Abas, FS 2008, 'Supervised Locally Linear Embedding in face recognition' Paper presented at IEEE- International Symposium on Biometrics and Security Technologies, ISBAST'08, Islamabad, Pakistan, 08/4/23 - 08/4/24, . https://doi.org/10.1109/ISBAST.2008.4547646

Supervised Locally Linear Embedding in face recognition. / Pang, Ying Han; Teoh, Beng Jin; Wong, Eng Kiong; Abas, Fazly Salleh.

2008. Paper presented at IEEE- International Symposium on Biometrics and Security Technologies, ISBAST'08, Islamabad, Pakistan.

Research output: Contribution to conferencePaper

TY - CONF

T1 - Supervised Locally Linear Embedding in face recognition

AU - Pang, Ying Han

AU - Teoh, Beng Jin

AU - Wong, Eng Kiong

AU - Abas, Fazly Salleh

PY - 2008/9/12

Y1 - 2008/9/12

N2 - Locally Linear Embedding (LLE), which has recently emerged as a powerful face feature descriptor, suffers from a limitation. That is class-specific information of data is lacked of during face analysis. Thus, we propose a supervised LLE technique, known as class-label Locally Linear Embedding (cLLE), to overcome the problem. cLLE is able to discover the nonlinearity of high-dimensional face data by minimizing the global reconstruction error of the set of all local neighbors in the data set. cLLE utilizes user class-specific information in neighborhoods selection and thus preserves the local neighborhoods. Since the locality preservation is correlated to the class discrimination, the proposed cLLE is expected superior to LLE in face recognition. Experimental results on three face databases: ORL, AR and Yale databases, demonstrate that the proposed technique obtains better recognition performance than PCA and LLE.

AB - Locally Linear Embedding (LLE), which has recently emerged as a powerful face feature descriptor, suffers from a limitation. That is class-specific information of data is lacked of during face analysis. Thus, we propose a supervised LLE technique, known as class-label Locally Linear Embedding (cLLE), to overcome the problem. cLLE is able to discover the nonlinearity of high-dimensional face data by minimizing the global reconstruction error of the set of all local neighbors in the data set. cLLE utilizes user class-specific information in neighborhoods selection and thus preserves the local neighborhoods. Since the locality preservation is correlated to the class discrimination, the proposed cLLE is expected superior to LLE in face recognition. Experimental results on three face databases: ORL, AR and Yale databases, demonstrate that the proposed technique obtains better recognition performance than PCA and LLE.

UR - http://www.scopus.com/inward/record.url?scp=51249113921&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=51249113921&partnerID=8YFLogxK

U2 - 10.1109/ISBAST.2008.4547646

DO - 10.1109/ISBAST.2008.4547646

M3 - Paper

ER -

Pang YH, Teoh BJ, Wong EK, Abas FS. Supervised Locally Linear Embedding in face recognition. 2008. Paper presented at IEEE- International Symposium on Biometrics and Security Technologies, ISBAST'08, Islamabad, Pakistan. https://doi.org/10.1109/ISBAST.2008.4547646