Face recognition based on neighbourhood discriminant preserving embedding

Andrew Beng Jin Teoh, Pang Ying Han

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

2 Citations (Scopus)

Abstract

Neighborhood Preserving Embedding (NPE) is an unsupervised linear dimensionality reduction technique which attempts to solve the "out of sample" problem in Locally Linear Embedding (LLE). This is done by introducing a linear transform matrix into LLE, and hence NPE can be perceived as a linear approximation to LLE. In this paper, we modify the original NPE for face recognition by embedding prior class information in the process of neighborhood selection. Intuitively, neighboring points are kept intact if they have the same class label, while avoid points of other classes from entering the neighborhood. We proved experimentally in three face databases, ie. ORL, PIE and FRGC, and with comparisons with other linear and non-linear feature extractors, the intuition underlying the inclusion of class information in NPE works out very advantageously for achieving high recognition performance.

Original languageEnglish
Title of host publication2008 10th International Conference on Control, Automation, Robotics and Vision, ICARCV 2008
Pages428-433
Number of pages6
DOIs
Publication statusPublished - 2008 Dec 1
Event2008 10th International Conference on Control, Automation, Robotics and Vision, ICARCV 2008 - Hanoi, Viet Nam
Duration: 2008 Dec 172008 Dec 20

Other

Other2008 10th International Conference on Control, Automation, Robotics and Vision, ICARCV 2008
CountryViet Nam
CityHanoi
Period08/12/1708/12/20

Fingerprint

Face recognition
Labels

All Science Journal Classification (ASJC) codes

  • Computer Vision and Pattern Recognition
  • Control and Systems Engineering
  • Electrical and Electronic Engineering

Cite this

Teoh, A. B. J., & Han, P. Y. (2008). Face recognition based on neighbourhood discriminant preserving embedding. In 2008 10th International Conference on Control, Automation, Robotics and Vision, ICARCV 2008 (pp. 428-433). [4795557] https://doi.org/10.1109/ICARCV.2008.4795557
Teoh, Andrew Beng Jin ; Han, Pang Ying. / Face recognition based on neighbourhood discriminant preserving embedding. 2008 10th International Conference on Control, Automation, Robotics and Vision, ICARCV 2008. 2008. pp. 428-433
@inproceedings{438a51ca19c945bd8754e036fd69887e,
title = "Face recognition based on neighbourhood discriminant preserving embedding",
abstract = "Neighborhood Preserving Embedding (NPE) is an unsupervised linear dimensionality reduction technique which attempts to solve the {"}out of sample{"} problem in Locally Linear Embedding (LLE). This is done by introducing a linear transform matrix into LLE, and hence NPE can be perceived as a linear approximation to LLE. In this paper, we modify the original NPE for face recognition by embedding prior class information in the process of neighborhood selection. Intuitively, neighboring points are kept intact if they have the same class label, while avoid points of other classes from entering the neighborhood. We proved experimentally in three face databases, ie. ORL, PIE and FRGC, and with comparisons with other linear and non-linear feature extractors, the intuition underlying the inclusion of class information in NPE works out very advantageously for achieving high recognition performance.",
author = "Teoh, {Andrew Beng Jin} and Han, {Pang Ying}",
year = "2008",
month = "12",
day = "1",
doi = "10.1109/ICARCV.2008.4795557",
language = "English",
isbn = "9781424422876",
pages = "428--433",
booktitle = "2008 10th International Conference on Control, Automation, Robotics and Vision, ICARCV 2008",

}

Teoh, ABJ & Han, PY 2008, Face recognition based on neighbourhood discriminant preserving embedding. in 2008 10th International Conference on Control, Automation, Robotics and Vision, ICARCV 2008., 4795557, pp. 428-433, 2008 10th International Conference on Control, Automation, Robotics and Vision, ICARCV 2008, Hanoi, Viet Nam, 08/12/17. https://doi.org/10.1109/ICARCV.2008.4795557

Face recognition based on neighbourhood discriminant preserving embedding. / Teoh, Andrew Beng Jin; Han, Pang Ying.

2008 10th International Conference on Control, Automation, Robotics and Vision, ICARCV 2008. 2008. p. 428-433 4795557.

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

TY - GEN

T1 - Face recognition based on neighbourhood discriminant preserving embedding

AU - Teoh, Andrew Beng Jin

AU - Han, Pang Ying

PY - 2008/12/1

Y1 - 2008/12/1

N2 - Neighborhood Preserving Embedding (NPE) is an unsupervised linear dimensionality reduction technique which attempts to solve the "out of sample" problem in Locally Linear Embedding (LLE). This is done by introducing a linear transform matrix into LLE, and hence NPE can be perceived as a linear approximation to LLE. In this paper, we modify the original NPE for face recognition by embedding prior class information in the process of neighborhood selection. Intuitively, neighboring points are kept intact if they have the same class label, while avoid points of other classes from entering the neighborhood. We proved experimentally in three face databases, ie. ORL, PIE and FRGC, and with comparisons with other linear and non-linear feature extractors, the intuition underlying the inclusion of class information in NPE works out very advantageously for achieving high recognition performance.

AB - Neighborhood Preserving Embedding (NPE) is an unsupervised linear dimensionality reduction technique which attempts to solve the "out of sample" problem in Locally Linear Embedding (LLE). This is done by introducing a linear transform matrix into LLE, and hence NPE can be perceived as a linear approximation to LLE. In this paper, we modify the original NPE for face recognition by embedding prior class information in the process of neighborhood selection. Intuitively, neighboring points are kept intact if they have the same class label, while avoid points of other classes from entering the neighborhood. We proved experimentally in three face databases, ie. ORL, PIE and FRGC, and with comparisons with other linear and non-linear feature extractors, the intuition underlying the inclusion of class information in NPE works out very advantageously for achieving high recognition performance.

UR - http://www.scopus.com/inward/record.url?scp=64549121923&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=64549121923&partnerID=8YFLogxK

U2 - 10.1109/ICARCV.2008.4795557

DO - 10.1109/ICARCV.2008.4795557

M3 - Conference contribution

SN - 9781424422876

SP - 428

EP - 433

BT - 2008 10th International Conference on Control, Automation, Robotics and Vision, ICARCV 2008

ER -

Teoh ABJ, Han PY. Face recognition based on neighbourhood discriminant preserving embedding. In 2008 10th International Conference on Control, Automation, Robotics and Vision, ICARCV 2008. 2008. p. 428-433. 4795557 https://doi.org/10.1109/ICARCV.2008.4795557