A study of aggregated 2D gabor features on appearance-based face recognition

King Hong Cheung; Jia You; Wai Kin Kong; Dapeng Zhang

A study of aggregated 2D gabor features on appearance-based face recognition

King Hong Cheung, Jia You, Wai Kin Kong, Dapeng Zhang

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

3 Citations (Scopus)

Abstract

Existing approaches to holistic appearance based face recognition require a high dimensional feature space to attain fruitful performance. We have proposed a relatively low feature dimensional scheme to deal with the face recognition problem. We use the aggregated responses of 2D Gabor filters to represent face images. We have investigated the effect of "duplicate" images and the effect of facial expressions. Our results show that the proposed method is more robust than the PCA-based method under varying facial expressions, especially in recognizing "duplicate" images.

Original language	English
Title of host publication	Proceedings - Third International Conference on Image and Graphics
Pages	310-313
Number of pages	4
Publication status	Published - 1 Dec 2004
Event	Proceedings - Third International Conference on Image and Graphics - Hong Kong, Hong Kong Duration: 18 Dec 2004 → 20 Dec 2004

Conference

Conference	Proceedings - Third International Conference on Image and Graphics
Country/Territory	Hong Kong
City	Hong Kong
Period	18/12/04 → 20/12/04

ASJC Scopus subject areas

General Engineering

More information

http://hdl.handle.net/10397/1171

Cite this

@inproceedings{ad1621215b6b4718b64935cedca6dbed,

title = "A study of aggregated 2D gabor features on appearance-based face recognition",

abstract = "Existing approaches to holistic appearance based face recognition require a high dimensional feature space to attain fruitful performance. We have proposed a relatively low feature dimensional scheme to deal with the face recognition problem. We use the aggregated responses of 2D Gabor filters to represent face images. We have investigated the effect of {"}duplicate{"} images and the effect of facial expressions. Our results show that the proposed method is more robust than the PCA-based method under varying facial expressions, especially in recognizing {"}duplicate{"} images.",

author = "Cheung, {King Hong} and Jia You and Kong, {Wai Kin} and Dapeng Zhang",

year = "2004",

month = dec,

day = "1",

language = "English",

isbn = "0769522440",

pages = "310--313",

booktitle = "Proceedings - Third International Conference on Image and Graphics",

note = "Proceedings - Third International Conference on Image and Graphics ; Conference date: 18-12-2004 Through 20-12-2004",

}

TY - GEN

T1 - A study of aggregated 2D gabor features on appearance-based face recognition

AU - Cheung, King Hong

AU - You, Jia

AU - Kong, Wai Kin

AU - Zhang, Dapeng

PY - 2004/12/1

Y1 - 2004/12/1

N2 - Existing approaches to holistic appearance based face recognition require a high dimensional feature space to attain fruitful performance. We have proposed a relatively low feature dimensional scheme to deal with the face recognition problem. We use the aggregated responses of 2D Gabor filters to represent face images. We have investigated the effect of "duplicate" images and the effect of facial expressions. Our results show that the proposed method is more robust than the PCA-based method under varying facial expressions, especially in recognizing "duplicate" images.

AB - Existing approaches to holistic appearance based face recognition require a high dimensional feature space to attain fruitful performance. We have proposed a relatively low feature dimensional scheme to deal with the face recognition problem. We use the aggregated responses of 2D Gabor filters to represent face images. We have investigated the effect of "duplicate" images and the effect of facial expressions. Our results show that the proposed method is more robust than the PCA-based method under varying facial expressions, especially in recognizing "duplicate" images.

UR - http://www.scopus.com/inward/record.url?scp=17044363461&partnerID=8YFLogxK

M3 - Conference article published in proceeding or book

SN - 0769522440

SP - 310

EP - 313

BT - Proceedings - Third International Conference on Image and Graphics

T2 - Proceedings - Third International Conference on Image and Graphics

Y2 - 18 December 2004 through 20 December 2004

ER -

A study of aggregated 2D gabor features on appearance-based face recognition

Abstract

Conference

ASJC Scopus subject areas

More information

Other files and links

Fingerprint

Cite this