Kernel-based regularized neighbourhood preserving embedding in face recognition

Pang Ying Han, Beng Jin Teoh

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

Abstract

Face images always have significant intra-class variations due to different poses, illuminations and facial expressions. These variations trigger substantial deviation from the linearity assumption of data structure, which is essential in formulating linear dimension reduction technique. In this paper, we present a kernel based regularized graph embedding dimension reduction technique, known as kernel-based Regularized Neighbourhood Preserving Embedding (KRNPE) to address this problem. KRNPE first exploits kernel function to unfold the nonlinear intrinsic facial data structure. Neighbourhood Preserving Embedding, a graph embedding based linear dimension reduction technique, is then regulated based on Adaptive Locality Preserving Regulation Model, established in [7] to enhance the locality preserving capability of the projection features, leading to better discriminating capability and generalization performance. Experimental results on PIE and FERET face databases validate the effectiveness of KRNPE.

Original languageEnglish
Title of host publicationProceedings of the 2012 7th IEEE Conference on Industrial Electronics and Applications, ICIEA 2012
Pages883-888
Number of pages6
DOIs
Publication statusPublished - 2012 Dec 1
Event2012 7th IEEE Conference on Industrial Electronics and Applications, ICIEA 2012 - Singapore, Singapore
Duration: 2012 Jul 182012 Jul 20

Publication series

NameProceedings of the 2012 7th IEEE Conference on Industrial Electronics and Applications, ICIEA 2012

Other

Other2012 7th IEEE Conference on Industrial Electronics and Applications, ICIEA 2012
CountrySingapore
CitySingapore
Period12/7/1812/7/20

Fingerprint

Face recognition
Data structures
Lighting

All Science Journal Classification (ASJC) codes

  • Electrical and Electronic Engineering

Cite this

Han, P. Y., & Teoh, B. J. (2012). Kernel-based regularized neighbourhood preserving embedding in face recognition. In Proceedings of the 2012 7th IEEE Conference on Industrial Electronics and Applications, ICIEA 2012 (pp. 883-888). [6360849] (Proceedings of the 2012 7th IEEE Conference on Industrial Electronics and Applications, ICIEA 2012). https://doi.org/10.1109/ICIEA.2012.6360849
Han, Pang Ying ; Teoh, Beng Jin. / Kernel-based regularized neighbourhood preserving embedding in face recognition. Proceedings of the 2012 7th IEEE Conference on Industrial Electronics and Applications, ICIEA 2012. 2012. pp. 883-888 (Proceedings of the 2012 7th IEEE Conference on Industrial Electronics and Applications, ICIEA 2012).
@inproceedings{10b7d2476c78405ab721d64cc4a8c7c2,
title = "Kernel-based regularized neighbourhood preserving embedding in face recognition",
abstract = "Face images always have significant intra-class variations due to different poses, illuminations and facial expressions. These variations trigger substantial deviation from the linearity assumption of data structure, which is essential in formulating linear dimension reduction technique. In this paper, we present a kernel based regularized graph embedding dimension reduction technique, known as kernel-based Regularized Neighbourhood Preserving Embedding (KRNPE) to address this problem. KRNPE first exploits kernel function to unfold the nonlinear intrinsic facial data structure. Neighbourhood Preserving Embedding, a graph embedding based linear dimension reduction technique, is then regulated based on Adaptive Locality Preserving Regulation Model, established in [7] to enhance the locality preserving capability of the projection features, leading to better discriminating capability and generalization performance. Experimental results on PIE and FERET face databases validate the effectiveness of KRNPE.",
author = "Han, {Pang Ying} and Teoh, {Beng Jin}",
year = "2012",
month = "12",
day = "1",
doi = "10.1109/ICIEA.2012.6360849",
language = "English",
isbn = "9781457721175",
series = "Proceedings of the 2012 7th IEEE Conference on Industrial Electronics and Applications, ICIEA 2012",
pages = "883--888",
booktitle = "Proceedings of the 2012 7th IEEE Conference on Industrial Electronics and Applications, ICIEA 2012",

}

Han, PY & Teoh, BJ 2012, Kernel-based regularized neighbourhood preserving embedding in face recognition. in Proceedings of the 2012 7th IEEE Conference on Industrial Electronics and Applications, ICIEA 2012., 6360849, Proceedings of the 2012 7th IEEE Conference on Industrial Electronics and Applications, ICIEA 2012, pp. 883-888, 2012 7th IEEE Conference on Industrial Electronics and Applications, ICIEA 2012, Singapore, Singapore, 12/7/18. https://doi.org/10.1109/ICIEA.2012.6360849

Kernel-based regularized neighbourhood preserving embedding in face recognition. / Han, Pang Ying; Teoh, Beng Jin.

Proceedings of the 2012 7th IEEE Conference on Industrial Electronics and Applications, ICIEA 2012. 2012. p. 883-888 6360849 (Proceedings of the 2012 7th IEEE Conference on Industrial Electronics and Applications, ICIEA 2012).

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

TY - GEN

T1 - Kernel-based regularized neighbourhood preserving embedding in face recognition

AU - Han, Pang Ying

AU - Teoh, Beng Jin

PY - 2012/12/1

Y1 - 2012/12/1

N2 - Face images always have significant intra-class variations due to different poses, illuminations and facial expressions. These variations trigger substantial deviation from the linearity assumption of data structure, which is essential in formulating linear dimension reduction technique. In this paper, we present a kernel based regularized graph embedding dimension reduction technique, known as kernel-based Regularized Neighbourhood Preserving Embedding (KRNPE) to address this problem. KRNPE first exploits kernel function to unfold the nonlinear intrinsic facial data structure. Neighbourhood Preserving Embedding, a graph embedding based linear dimension reduction technique, is then regulated based on Adaptive Locality Preserving Regulation Model, established in [7] to enhance the locality preserving capability of the projection features, leading to better discriminating capability and generalization performance. Experimental results on PIE and FERET face databases validate the effectiveness of KRNPE.

AB - Face images always have significant intra-class variations due to different poses, illuminations and facial expressions. These variations trigger substantial deviation from the linearity assumption of data structure, which is essential in formulating linear dimension reduction technique. In this paper, we present a kernel based regularized graph embedding dimension reduction technique, known as kernel-based Regularized Neighbourhood Preserving Embedding (KRNPE) to address this problem. KRNPE first exploits kernel function to unfold the nonlinear intrinsic facial data structure. Neighbourhood Preserving Embedding, a graph embedding based linear dimension reduction technique, is then regulated based on Adaptive Locality Preserving Regulation Model, established in [7] to enhance the locality preserving capability of the projection features, leading to better discriminating capability and generalization performance. Experimental results on PIE and FERET face databases validate the effectiveness of KRNPE.

UR - http://www.scopus.com/inward/record.url?scp=84871684702&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84871684702&partnerID=8YFLogxK

U2 - 10.1109/ICIEA.2012.6360849

DO - 10.1109/ICIEA.2012.6360849

M3 - Conference contribution

AN - SCOPUS:84871684702

SN - 9781457721175

T3 - Proceedings of the 2012 7th IEEE Conference on Industrial Electronics and Applications, ICIEA 2012

SP - 883

EP - 888

BT - Proceedings of the 2012 7th IEEE Conference on Industrial Electronics and Applications, ICIEA 2012

ER -

Han PY, Teoh BJ. Kernel-based regularized neighbourhood preserving embedding in face recognition. In Proceedings of the 2012 7th IEEE Conference on Industrial Electronics and Applications, ICIEA 2012. 2012. p. 883-888. 6360849. (Proceedings of the 2012 7th IEEE Conference on Industrial Electronics and Applications, ICIEA 2012). https://doi.org/10.1109/ICIEA.2012.6360849