Gabor Surface Feature for face recognition

Ke Yan; Youbin Chen; Dapeng Zhang

doi:10.1109/ACPR.2011.6166553

Gabor Surface Feature for face recognition

Ke Yan, Youbin Chen, Dapeng Zhang

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

25 Citations (Scopus)

Abstract

Gabor filters can extract multi-orientation and multiscale features from face images. Researchers have designed different ways to use the magnitude of the filtered results for face recognition: Gabor Fisher classifier exploited only the magnitude information of Gabor magnitude pictures (GMPs); Local Gabor Binary Pattern uses only the gradient information. In this paper, we regard GMPs as smooth surfaces. By completely describing the shape of GMPs, we get a face representation method called Gabor Surface Feature (GSF). First, we compute the magnitude, 1stand 2ndderivatives of GMPs, then binarize them and transform them into decimal values. Finally we construct joint histograms and use subspace methods for classification. Experiments on FERET, ORL and FRGC 1.0.4 database show the effectiveness of GSF.

Original language	English
Title of host publication	1st Asian Conference on Pattern Recognition, ACPR 2011
Pages	288-292
Number of pages	5
DOIs	https://doi.org/10.1109/ACPR.2011.6166553
Publication status	Published - 1 Dec 2011
Event	1st Asian Conference on Pattern Recognition, ACPR 2011 - Beijing, China Duration: 28 Nov 2011 → 28 Nov 2011

Conference

Conference	1st Asian Conference on Pattern Recognition, ACPR 2011
Country/Territory	China
City	Beijing
Period	28/11/11 → 28/11/11

Keywords

face recognition
feature extraction
Gabor
Gabor surface feature
histogram

ASJC Scopus subject areas

Computer Vision and Pattern Recognition

More information

10.1109/ACPR.2011.6166553

Cite this

@inproceedings{dbad50f2314e4c759e8f83f5715cd521,

title = "Gabor Surface Feature for face recognition",

abstract = "Gabor filters can extract multi-orientation and multiscale features from face images. Researchers have designed different ways to use the magnitude of the filtered results for face recognition: Gabor Fisher classifier exploited only the magnitude information of Gabor magnitude pictures (GMPs); Local Gabor Binary Pattern uses only the gradient information. In this paper, we regard GMPs as smooth surfaces. By completely describing the shape of GMPs, we get a face representation method called Gabor Surface Feature (GSF). First, we compute the magnitude, 1stand 2ndderivatives of GMPs, then binarize them and transform them into decimal values. Finally we construct joint histograms and use subspace methods for classification. Experiments on FERET, ORL and FRGC 1.0.4 database show the effectiveness of GSF.",

keywords = "face recognition, feature extraction, Gabor, Gabor surface feature, histogram",

author = "Ke Yan and Youbin Chen and Dapeng Zhang",

year = "2011",

month = dec,

day = "1",

doi = "10.1109/ACPR.2011.6166553",

language = "English",

isbn = "9781457701221",

pages = "288--292",

booktitle = "1st Asian Conference on Pattern Recognition, ACPR 2011",

note = "1st Asian Conference on Pattern Recognition, ACPR 2011 ; Conference date: 28-11-2011 Through 28-11-2011",

}

TY - GEN

T1 - Gabor Surface Feature for face recognition

AU - Yan, Ke

AU - Chen, Youbin

AU - Zhang, Dapeng

PY - 2011/12/1

Y1 - 2011/12/1

N2 - Gabor filters can extract multi-orientation and multiscale features from face images. Researchers have designed different ways to use the magnitude of the filtered results for face recognition: Gabor Fisher classifier exploited only the magnitude information of Gabor magnitude pictures (GMPs); Local Gabor Binary Pattern uses only the gradient information. In this paper, we regard GMPs as smooth surfaces. By completely describing the shape of GMPs, we get a face representation method called Gabor Surface Feature (GSF). First, we compute the magnitude, 1stand 2ndderivatives of GMPs, then binarize them and transform them into decimal values. Finally we construct joint histograms and use subspace methods for classification. Experiments on FERET, ORL and FRGC 1.0.4 database show the effectiveness of GSF.

AB - Gabor filters can extract multi-orientation and multiscale features from face images. Researchers have designed different ways to use the magnitude of the filtered results for face recognition: Gabor Fisher classifier exploited only the magnitude information of Gabor magnitude pictures (GMPs); Local Gabor Binary Pattern uses only the gradient information. In this paper, we regard GMPs as smooth surfaces. By completely describing the shape of GMPs, we get a face representation method called Gabor Surface Feature (GSF). First, we compute the magnitude, 1stand 2ndderivatives of GMPs, then binarize them and transform them into decimal values. Finally we construct joint histograms and use subspace methods for classification. Experiments on FERET, ORL and FRGC 1.0.4 database show the effectiveness of GSF.

KW - face recognition

KW - feature extraction

KW - Gabor

KW - Gabor surface feature

KW - histogram

UR - http://www.scopus.com/inward/record.url?scp=84862867099&partnerID=8YFLogxK

U2 - 10.1109/ACPR.2011.6166553

DO - 10.1109/ACPR.2011.6166553

M3 - Conference article published in proceeding or book

SN - 9781457701221

SP - 288

EP - 292

BT - 1st Asian Conference on Pattern Recognition, ACPR 2011

T2 - 1st Asian Conference on Pattern Recognition, ACPR 2011

Y2 - 28 November 2011 through 28 November 2011

ER -

Gabor Surface Feature for face recognition

Abstract

Conference

Keywords

ASJC Scopus subject areas

More information

Other files and links

Cite this