Eigenvector weighting function in face recognition

Beng Jin Teoh, Pang Ying Han, Lim Heng Siong

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

Graph-based subspace learning is a class of dimensionality reduction technique in face recognition. The technique reveals the local manifold structure of face data that hidden in the image space via a linear projection. However, the real world face data may be too complex to measure due to both external imaging noises and the intra-class variations of the face images. Hence, features which are extracted by the graph-based technique could be noisy. An appropriate weight should be imposed to the data features for better data discrimination. In this paper, a piecewise weighting function, known as Eigenvector Weighting Function (EWF), is proposed and implemented in two graph based subspace learning techniques, namely Locality Preserving Projection and Neighbourhood Preserving Embedding. Specifically, the computed projection subspace of the learning approach is decomposed into three partitions: a subspace due to intra-class variations, an intrinsic face subspace, and a subspace which is attributed to imaging noises. Projected data features are weighted differently in these subspaces to emphasize the intrinsic face subspace while penalizing the other two subspaces. Experiments on FERET and FRGC databases are conducted to show the promising performance of the proposed technique.

Original languageEnglish
Article number521935
JournalDiscrete Dynamics in Nature and Society
Volume2011
DOIs
Publication statusPublished - 2011 Apr 26

Fingerprint

Weighting Function
Face recognition
Face Recognition
Eigenvalues and eigenfunctions
Eigenvector
Subspace
Imaging techniques
Face
Experiments
Graph in graph theory
Imaging
Projection
Linear Projection
Image Space
Dimensionality Reduction
Locality
Discrimination
Partition

All Science Journal Classification (ASJC) codes

  • Modelling and Simulation

Cite this

@article{d63c374bfb7646aa8e34598281264ee7,
title = "Eigenvector weighting function in face recognition",
abstract = "Graph-based subspace learning is a class of dimensionality reduction technique in face recognition. The technique reveals the local manifold structure of face data that hidden in the image space via a linear projection. However, the real world face data may be too complex to measure due to both external imaging noises and the intra-class variations of the face images. Hence, features which are extracted by the graph-based technique could be noisy. An appropriate weight should be imposed to the data features for better data discrimination. In this paper, a piecewise weighting function, known as Eigenvector Weighting Function (EWF), is proposed and implemented in two graph based subspace learning techniques, namely Locality Preserving Projection and Neighbourhood Preserving Embedding. Specifically, the computed projection subspace of the learning approach is decomposed into three partitions: a subspace due to intra-class variations, an intrinsic face subspace, and a subspace which is attributed to imaging noises. Projected data features are weighted differently in these subspaces to emphasize the intrinsic face subspace while penalizing the other two subspaces. Experiments on FERET and FRGC databases are conducted to show the promising performance of the proposed technique.",
author = "Teoh, {Beng Jin} and Han, {Pang Ying} and Siong, {Lim Heng}",
year = "2011",
month = "4",
day = "26",
doi = "10.1155/2011/521935",
language = "English",
volume = "2011",
journal = "Discrete Dynamics in Nature and Society",
issn = "1026-0226",
publisher = "Hindawi Publishing Corporation",

}

Eigenvector weighting function in face recognition. / Teoh, Beng Jin; Han, Pang Ying; Siong, Lim Heng.

In: Discrete Dynamics in Nature and Society, Vol. 2011, 521935, 26.04.2011.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Eigenvector weighting function in face recognition

AU - Teoh, Beng Jin

AU - Han, Pang Ying

AU - Siong, Lim Heng

PY - 2011/4/26

Y1 - 2011/4/26

N2 - Graph-based subspace learning is a class of dimensionality reduction technique in face recognition. The technique reveals the local manifold structure of face data that hidden in the image space via a linear projection. However, the real world face data may be too complex to measure due to both external imaging noises and the intra-class variations of the face images. Hence, features which are extracted by the graph-based technique could be noisy. An appropriate weight should be imposed to the data features for better data discrimination. In this paper, a piecewise weighting function, known as Eigenvector Weighting Function (EWF), is proposed and implemented in two graph based subspace learning techniques, namely Locality Preserving Projection and Neighbourhood Preserving Embedding. Specifically, the computed projection subspace of the learning approach is decomposed into three partitions: a subspace due to intra-class variations, an intrinsic face subspace, and a subspace which is attributed to imaging noises. Projected data features are weighted differently in these subspaces to emphasize the intrinsic face subspace while penalizing the other two subspaces. Experiments on FERET and FRGC databases are conducted to show the promising performance of the proposed technique.

AB - Graph-based subspace learning is a class of dimensionality reduction technique in face recognition. The technique reveals the local manifold structure of face data that hidden in the image space via a linear projection. However, the real world face data may be too complex to measure due to both external imaging noises and the intra-class variations of the face images. Hence, features which are extracted by the graph-based technique could be noisy. An appropriate weight should be imposed to the data features for better data discrimination. In this paper, a piecewise weighting function, known as Eigenvector Weighting Function (EWF), is proposed and implemented in two graph based subspace learning techniques, namely Locality Preserving Projection and Neighbourhood Preserving Embedding. Specifically, the computed projection subspace of the learning approach is decomposed into three partitions: a subspace due to intra-class variations, an intrinsic face subspace, and a subspace which is attributed to imaging noises. Projected data features are weighted differently in these subspaces to emphasize the intrinsic face subspace while penalizing the other two subspaces. Experiments on FERET and FRGC databases are conducted to show the promising performance of the proposed technique.

UR - http://www.scopus.com/inward/record.url?scp=79955037037&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79955037037&partnerID=8YFLogxK

U2 - 10.1155/2011/521935

DO - 10.1155/2011/521935

M3 - Article

AN - SCOPUS:79955037037

VL - 2011

JO - Discrete Dynamics in Nature and Society

JF - Discrete Dynamics in Nature and Society

SN - 1026-0226

M1 - 521935

ER -