Combining local face image features for identity verification

Research output: Contribution to journalArticle

11 Citations (Scopus)

Abstract

With an aim of extracting robust facial features under pose variations, this paper presents two directional projections corresponding to extraction of vertical and horizontal local face image features. The matching scores computed from both horizontal and vertical features are subsequently fused at score level via an extreme learning machine that optimizes the total error rate for performance enhancement. In order to benchmark the performance, both the feature extraction and fusion results are compared with that of popular face recognition methods such as principal components analysis and linear discriminant analysis in terms of equal error rate and CPU time. Our empirical experiments using four data sets show encouraging results under considerable horizontal pose variations.

Original languageEnglish
Pages (from-to)2452-2463
Number of pages12
JournalNeurocomputing
Volume74
Issue number16
DOIs
Publication statusPublished - 2011 Sep 1

Fingerprint

Benchmarking
Discriminant Analysis
Principal Component Analysis
Discriminant analysis
Face recognition
Principal component analysis
Program processors
Learning systems
Feature extraction
Fusion reactions
Experiments
Facial Recognition
Machine Learning
Datasets

All Science Journal Classification (ASJC) codes

  • Computer Science Applications
  • Cognitive Neuroscience
  • Artificial Intelligence

Cite this

@article{b9d2d239492b4d2aaff26eab420c6d0e,
title = "Combining local face image features for identity verification",
abstract = "With an aim of extracting robust facial features under pose variations, this paper presents two directional projections corresponding to extraction of vertical and horizontal local face image features. The matching scores computed from both horizontal and vertical features are subsequently fused at score level via an extreme learning machine that optimizes the total error rate for performance enhancement. In order to benchmark the performance, both the feature extraction and fusion results are compared with that of popular face recognition methods such as principal components analysis and linear discriminant analysis in terms of equal error rate and CPU time. Our empirical experiments using four data sets show encouraging results under considerable horizontal pose variations.",
author = "Oh, {Beom Seok} and Toh, {Kar Ann} and Teoh, {Beng Jin} and Jaihie Kim",
year = "2011",
month = "9",
day = "1",
doi = "10.1016/j.neucom.2010.12.036",
language = "English",
volume = "74",
pages = "2452--2463",
journal = "Neurocomputing",
issn = "0925-2312",
publisher = "Elsevier",
number = "16",

}

Combining local face image features for identity verification. / Oh, Beom Seok; Toh, Kar Ann; Teoh, Beng Jin; Kim, Jaihie.

In: Neurocomputing, Vol. 74, No. 16, 01.09.2011, p. 2452-2463.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Combining local face image features for identity verification

AU - Oh, Beom Seok

AU - Toh, Kar Ann

AU - Teoh, Beng Jin

AU - Kim, Jaihie

PY - 2011/9/1

Y1 - 2011/9/1

N2 - With an aim of extracting robust facial features under pose variations, this paper presents two directional projections corresponding to extraction of vertical and horizontal local face image features. The matching scores computed from both horizontal and vertical features are subsequently fused at score level via an extreme learning machine that optimizes the total error rate for performance enhancement. In order to benchmark the performance, both the feature extraction and fusion results are compared with that of popular face recognition methods such as principal components analysis and linear discriminant analysis in terms of equal error rate and CPU time. Our empirical experiments using four data sets show encouraging results under considerable horizontal pose variations.

AB - With an aim of extracting robust facial features under pose variations, this paper presents two directional projections corresponding to extraction of vertical and horizontal local face image features. The matching scores computed from both horizontal and vertical features are subsequently fused at score level via an extreme learning machine that optimizes the total error rate for performance enhancement. In order to benchmark the performance, both the feature extraction and fusion results are compared with that of popular face recognition methods such as principal components analysis and linear discriminant analysis in terms of equal error rate and CPU time. Our empirical experiments using four data sets show encouraging results under considerable horizontal pose variations.

UR - http://www.scopus.com/inward/record.url?scp=80051577109&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=80051577109&partnerID=8YFLogxK

U2 - 10.1016/j.neucom.2010.12.036

DO - 10.1016/j.neucom.2010.12.036

M3 - Article

VL - 74

SP - 2452

EP - 2463

JO - Neurocomputing

JF - Neurocomputing

SN - 0925-2312

IS - 16

ER -