Unconstrained face verification with a dual-layer block-based metric learning

Siew Chin Chong, Beng Jin Teoh, Thian Song Ong

Research output: Contribution to journalArticle

Abstract

In this paper, a dual-layer block-based metric learning technique is proposed to better discriminate the face image pairs and accelerate the overall verification process under the unconstrained environment. The input images are processed as blocks to provide a richer base of face features. Our proposed method is formed by two layers, in which the first layer assists in extracting the compact block-based descriptors without the existence of full class label information and to refine the within-class and between-class scatter matrices while the second layer integrates the face descriptors of all blocks. The proposed scheme has computational advantage over the single metric learning method while it exploits the correlations among the multiple metrics from different descriptors. The performance of our proposed method is evaluated on the Labeled Faces in the Wild database and achieves an improved performance when compared with the state-of-the-art methods in terms of verification rate and computation time.

Original languageEnglish
Pages (from-to)1703-1719
Number of pages17
JournalMultimedia Tools and Applications
Volume76
Issue number2
DOIs
Publication statusPublished - 2017 Jan 1

Fingerprint

Labels

All Science Journal Classification (ASJC) codes

  • Software
  • Media Technology
  • Hardware and Architecture
  • Computer Networks and Communications

Cite this

@article{8da11905d5574c34b45019509ea17ce2,
title = "Unconstrained face verification with a dual-layer block-based metric learning",
abstract = "In this paper, a dual-layer block-based metric learning technique is proposed to better discriminate the face image pairs and accelerate the overall verification process under the unconstrained environment. The input images are processed as blocks to provide a richer base of face features. Our proposed method is formed by two layers, in which the first layer assists in extracting the compact block-based descriptors without the existence of full class label information and to refine the within-class and between-class scatter matrices while the second layer integrates the face descriptors of all blocks. The proposed scheme has computational advantage over the single metric learning method while it exploits the correlations among the multiple metrics from different descriptors. The performance of our proposed method is evaluated on the Labeled Faces in the Wild database and achieves an improved performance when compared with the state-of-the-art methods in terms of verification rate and computation time.",
author = "Chong, {Siew Chin} and Teoh, {Beng Jin} and Ong, {Thian Song}",
year = "2017",
month = "1",
day = "1",
doi = "10.1007/s11042-015-3120-5",
language = "English",
volume = "76",
pages = "1703--1719",
journal = "Multimedia Tools and Applications",
issn = "1380-7501",
publisher = "Springer Netherlands",
number = "2",

}

Unconstrained face verification with a dual-layer block-based metric learning. / Chong, Siew Chin; Teoh, Beng Jin; Ong, Thian Song.

In: Multimedia Tools and Applications, Vol. 76, No. 2, 01.01.2017, p. 1703-1719.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Unconstrained face verification with a dual-layer block-based metric learning

AU - Chong, Siew Chin

AU - Teoh, Beng Jin

AU - Ong, Thian Song

PY - 2017/1/1

Y1 - 2017/1/1

N2 - In this paper, a dual-layer block-based metric learning technique is proposed to better discriminate the face image pairs and accelerate the overall verification process under the unconstrained environment. The input images are processed as blocks to provide a richer base of face features. Our proposed method is formed by two layers, in which the first layer assists in extracting the compact block-based descriptors without the existence of full class label information and to refine the within-class and between-class scatter matrices while the second layer integrates the face descriptors of all blocks. The proposed scheme has computational advantage over the single metric learning method while it exploits the correlations among the multiple metrics from different descriptors. The performance of our proposed method is evaluated on the Labeled Faces in the Wild database and achieves an improved performance when compared with the state-of-the-art methods in terms of verification rate and computation time.

AB - In this paper, a dual-layer block-based metric learning technique is proposed to better discriminate the face image pairs and accelerate the overall verification process under the unconstrained environment. The input images are processed as blocks to provide a richer base of face features. Our proposed method is formed by two layers, in which the first layer assists in extracting the compact block-based descriptors without the existence of full class label information and to refine the within-class and between-class scatter matrices while the second layer integrates the face descriptors of all blocks. The proposed scheme has computational advantage over the single metric learning method while it exploits the correlations among the multiple metrics from different descriptors. The performance of our proposed method is evaluated on the Labeled Faces in the Wild database and achieves an improved performance when compared with the state-of-the-art methods in terms of verification rate and computation time.

UR - http://www.scopus.com/inward/record.url?scp=84951764936&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84951764936&partnerID=8YFLogxK

U2 - 10.1007/s11042-015-3120-5

DO - 10.1007/s11042-015-3120-5

M3 - Article

VL - 76

SP - 1703

EP - 1719

JO - Multimedia Tools and Applications

JF - Multimedia Tools and Applications

SN - 1380-7501

IS - 2

ER -