Hierarchical multi-label framework for robust face recognition

Lingfeng Zhang; Pengfei Dou; Shishir K. Shah; Ioannis A. Kakadiaris

doi:10.1109/ICB.2015.7139086

Hierarchical multi-label framework for robust face recognition

Lingfeng Zhang, Pengfei Dou, Shishir K. Shah, Ioannis A. Kakadiaris

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

4 Scopus citations

Abstract

In this paper, we propose a patch based face recognition framework. First, a face image is iteratively divided into multi-level patches and assigned hierarchical labels. Second, local classifiers are built to learn the local prediction of each patch. Third, the hierarchical relationships defined between local patches are used to obtain the global prediction of each patch. We develop three ways to learn the global prediction: majority voting, l₁-regularized weighting, and decision rule. Last, the global predictions of different levels are combined as the final prediction. Experimental results on different face recognition tasks demonstrate the effectiveness of our method.

Original language	English (US)
Title of host publication	Proceedings of 2015 International Conference on Biometrics, ICB 2015
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	127-134
Number of pages	8
ISBN (Electronic)	9781479978243
DOIs	https://doi.org/10.1109/ICB.2015.7139086
State	Published - Jun 29 2015
Event	8th IAPR International Conference on Biometrics, ICB 2015 - Phuket, Thailand Duration: May 19 2015 → May 22 2015

Publication series

Name	Proceedings of 2015 International Conference on Biometrics, ICB 2015

Conference

Conference	8th IAPR International Conference on Biometrics, ICB 2015
Country/Territory	Thailand
City	Phuket
Period	5/19/15 → 5/22/15

ASJC Scopus subject areas

Biotechnology
Computer Science Applications
Biomedical Engineering
Safety, Risk, Reliability and Quality

Access to Document

10.1109/ICB.2015.7139086

Cite this

Zhang, L., Dou, P., Shah, S. K., & Kakadiaris, I. A. (2015). Hierarchical multi-label framework for robust face recognition. In Proceedings of 2015 International Conference on Biometrics, ICB 2015 (pp. 127-134). Article 7139086 (Proceedings of 2015 International Conference on Biometrics, ICB 2015). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICB.2015.7139086

Hierarchical multi-label framework for robust face recognition. / Zhang, Lingfeng; Dou, Pengfei; Shah, Shishir K. et al.
Proceedings of 2015 International Conference on Biometrics, ICB 2015. Institute of Electrical and Electronics Engineers Inc., 2015. p. 127-134 7139086 (Proceedings of 2015 International Conference on Biometrics, ICB 2015).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Zhang, L, Dou, P, Shah, SK & Kakadiaris, IA 2015, Hierarchical multi-label framework for robust face recognition. in Proceedings of 2015 International Conference on Biometrics, ICB 2015., 7139086, Proceedings of 2015 International Conference on Biometrics, ICB 2015, Institute of Electrical and Electronics Engineers Inc., pp. 127-134, 8th IAPR International Conference on Biometrics, ICB 2015, Phuket, Thailand, 5/19/15. https://doi.org/10.1109/ICB.2015.7139086

@inproceedings{e4f484a01df741db9f26ff494f48e312,

title = "Hierarchical multi-label framework for robust face recognition",

abstract = "In this paper, we propose a patch based face recognition framework. First, a face image is iteratively divided into multi-level patches and assigned hierarchical labels. Second, local classifiers are built to learn the local prediction of each patch. Third, the hierarchical relationships defined between local patches are used to obtain the global prediction of each patch. We develop three ways to learn the global prediction: majority voting, l1-regularized weighting, and decision rule. Last, the global predictions of different levels are combined as the final prediction. Experimental results on different face recognition tasks demonstrate the effectiveness of our method.",

author = "Lingfeng Zhang and Pengfei Dou and Shah, {Shishir K.} and Kakadiaris, {Ioannis A.}",

note = "Publisher Copyright: {\textcopyright} 2015 IEEE.; 8th IAPR International Conference on Biometrics, ICB 2015 ; Conference date: 19-05-2015 Through 22-05-2015",

year = "2015",

month = jun,

day = "29",

doi = "10.1109/ICB.2015.7139086",

language = "English (US)",

series = "Proceedings of 2015 International Conference on Biometrics, ICB 2015",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "127--134",

booktitle = "Proceedings of 2015 International Conference on Biometrics, ICB 2015",

address = "United States",

}

TY - GEN

T1 - Hierarchical multi-label framework for robust face recognition

AU - Zhang, Lingfeng

AU - Dou, Pengfei

AU - Shah, Shishir K.

AU - Kakadiaris, Ioannis A.

PY - 2015/6/29

Y1 - 2015/6/29

N2 - In this paper, we propose a patch based face recognition framework. First, a face image is iteratively divided into multi-level patches and assigned hierarchical labels. Second, local classifiers are built to learn the local prediction of each patch. Third, the hierarchical relationships defined between local patches are used to obtain the global prediction of each patch. We develop three ways to learn the global prediction: majority voting, l1-regularized weighting, and decision rule. Last, the global predictions of different levels are combined as the final prediction. Experimental results on different face recognition tasks demonstrate the effectiveness of our method.

AB - In this paper, we propose a patch based face recognition framework. First, a face image is iteratively divided into multi-level patches and assigned hierarchical labels. Second, local classifiers are built to learn the local prediction of each patch. Third, the hierarchical relationships defined between local patches are used to obtain the global prediction of each patch. We develop three ways to learn the global prediction: majority voting, l1-regularized weighting, and decision rule. Last, the global predictions of different levels are combined as the final prediction. Experimental results on different face recognition tasks demonstrate the effectiveness of our method.

UR - http://www.scopus.com/inward/record.url?scp=84943234246&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84943234246&partnerID=8YFLogxK

U2 - 10.1109/ICB.2015.7139086

DO - 10.1109/ICB.2015.7139086

M3 - Conference contribution

AN - SCOPUS:84943234246

T3 - Proceedings of 2015 International Conference on Biometrics, ICB 2015

SP - 127

EP - 134

BT - Proceedings of 2015 International Conference on Biometrics, ICB 2015

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 8th IAPR International Conference on Biometrics, ICB 2015

Y2 - 19 May 2015 through 22 May 2015

ER -

Hierarchical multi-label framework for robust face recognition

Abstract

Publication series

Conference

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this