Fusion of structured projections for cancelable face identity verification

Beom Seok Oh, Kar Ann Toh

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

Abstract

This work proposes a structured random projection via feature weighting for cancelable identity verification. Essentially, projected facial features are weighted based on their discrimination capability prior to a matching process. In order to conceal the face identity, an averaging over several templates with different transformations is performed. Finally, several cancelable templates extracted from partial face images are fused at score level via a total error rate minimization. Our empirical experiments on two experimental scenarios using AR, FERET and Sheffield databases show that the proposed method consistently outperforms competing state-of-the-art un-supervised methods in terms of verification accuracy.

Original languageEnglish
Title of host publication2011 International Joint Conference on Biometrics, IJCB 2011
DOIs
Publication statusPublished - 2011 Dec 1
Event2011 International Joint Conference on Biometrics, IJCB 2011 - Washington, DC, United States
Duration: 2011 Oct 112011 Oct 13

Other

Other2011 International Joint Conference on Biometrics, IJCB 2011
CountryUnited States
CityWashington, DC
Period11/10/1111/10/13

Fingerprint

Databases

All Science Journal Classification (ASJC) codes

  • Biotechnology

Cite this

Oh, B. S., & Toh, K. A. (2011). Fusion of structured projections for cancelable face identity verification. In 2011 International Joint Conference on Biometrics, IJCB 2011 [6117588] https://doi.org/10.1109/IJCB.2011.6117588
Oh, Beom Seok ; Toh, Kar Ann. / Fusion of structured projections for cancelable face identity verification. 2011 International Joint Conference on Biometrics, IJCB 2011. 2011.
@inproceedings{f0b59cd656b448f6aed13ee4c5f0aa2f,
title = "Fusion of structured projections for cancelable face identity verification",
abstract = "This work proposes a structured random projection via feature weighting for cancelable identity verification. Essentially, projected facial features are weighted based on their discrimination capability prior to a matching process. In order to conceal the face identity, an averaging over several templates with different transformations is performed. Finally, several cancelable templates extracted from partial face images are fused at score level via a total error rate minimization. Our empirical experiments on two experimental scenarios using AR, FERET and Sheffield databases show that the proposed method consistently outperforms competing state-of-the-art un-supervised methods in terms of verification accuracy.",
author = "Oh, {Beom Seok} and Toh, {Kar Ann}",
year = "2011",
month = "12",
day = "1",
doi = "10.1109/IJCB.2011.6117588",
language = "English",
isbn = "9781457713583",
booktitle = "2011 International Joint Conference on Biometrics, IJCB 2011",

}

Oh, BS & Toh, KA 2011, Fusion of structured projections for cancelable face identity verification. in 2011 International Joint Conference on Biometrics, IJCB 2011., 6117588, 2011 International Joint Conference on Biometrics, IJCB 2011, Washington, DC, United States, 11/10/11. https://doi.org/10.1109/IJCB.2011.6117588

Fusion of structured projections for cancelable face identity verification. / Oh, Beom Seok; Toh, Kar Ann.

2011 International Joint Conference on Biometrics, IJCB 2011. 2011. 6117588.

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

TY - GEN

T1 - Fusion of structured projections for cancelable face identity verification

AU - Oh, Beom Seok

AU - Toh, Kar Ann

PY - 2011/12/1

Y1 - 2011/12/1

N2 - This work proposes a structured random projection via feature weighting for cancelable identity verification. Essentially, projected facial features are weighted based on their discrimination capability prior to a matching process. In order to conceal the face identity, an averaging over several templates with different transformations is performed. Finally, several cancelable templates extracted from partial face images are fused at score level via a total error rate minimization. Our empirical experiments on two experimental scenarios using AR, FERET and Sheffield databases show that the proposed method consistently outperforms competing state-of-the-art un-supervised methods in terms of verification accuracy.

AB - This work proposes a structured random projection via feature weighting for cancelable identity verification. Essentially, projected facial features are weighted based on their discrimination capability prior to a matching process. In order to conceal the face identity, an averaging over several templates with different transformations is performed. Finally, several cancelable templates extracted from partial face images are fused at score level via a total error rate minimization. Our empirical experiments on two experimental scenarios using AR, FERET and Sheffield databases show that the proposed method consistently outperforms competing state-of-the-art un-supervised methods in terms of verification accuracy.

UR - http://www.scopus.com/inward/record.url?scp=84856094263&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84856094263&partnerID=8YFLogxK

U2 - 10.1109/IJCB.2011.6117588

DO - 10.1109/IJCB.2011.6117588

M3 - Conference contribution

SN - 9781457713583

BT - 2011 International Joint Conference on Biometrics, IJCB 2011

ER -

Oh BS, Toh KA. Fusion of structured projections for cancelable face identity verification. In 2011 International Joint Conference on Biometrics, IJCB 2011. 2011. 6117588 https://doi.org/10.1109/IJCB.2011.6117588