Robust face recognition under varying illumination and occlusion considering structured sparsity

Xingjie Wei, Chang Tsun Li, Yongjian Hu

Research output: Book chapter/Published conference paperConference paper

19 Citations (Scopus)

Abstract

A large amount of work has been done over the past decades in face recognition (FR). Most of them deal with uncontrolled variations such as changes in illumination, pose, expression and occlusion individually. However, limited work focuses on simultaneously handling multiple variations. In real-world environment, uncontrolled variations usually coexist. FR approaches which are robust to one kind of variation may fail to deal with another. In this paper, we propose an approach considering structured sparsity to deal with the illumination changes and occlusion at the same time. Our approach represents a face image taking into account that the face images usually lie in the structured union of subspaces in a high dimensional feature space. Considering the spatial continuity of the occlusion, we propose a cluster occlusion dictionary for occlusion modelling. In addition, a discriminative feature is embedded in our model to correct the illumination effect. This enables our approach to handle images that lie outside the illumination subspace spanned by the training set. Experimental results on public face databases show that the proposed approach is very robust to large illumination changes and occlusion.
Original languageEnglish
Title of host publicationProceedings of the 2012 International Conference on Digital Image Computing: Techniques and Applications
Place of PublicationUnited States
PublisherIEEE, Institute of Electrical and Electronics Engineers
Pages1-7
Number of pages7
ISBN (Electronic)9781467321815
DOIs
Publication statusPublished - 2012
Event2012 International Conference on Digital Image Computing: Techniques and Applications (DICTA) - Esplanade Hotel, Fremantle, Australia
Duration: 03 Dec 201205 Dec 2012

Conference

Conference2012 International Conference on Digital Image Computing: Techniques and Applications (DICTA)
CountryAustralia
CityFremantle
Period03/12/1205/12/12
OtherThe International Conference on Digital Image Computing: Techniques and Applications (DICTA) is the main Australian Conference on computer vision, image processing, pattern recognition, and related areas. DICTA was established as a biannual conference in 1991 and became an annual event in 2007. It is the premiere conference of the Australian Pattern Recognition Society (APRS).

Fingerprint

Face recognition
Lighting
Glossaries

Cite this

Wei, X., Li, C. T., & Hu, Y. (2012). Robust face recognition under varying illumination and occlusion considering structured sparsity. In Proceedings of the 2012 International Conference on Digital Image Computing: Techniques and Applications (pp. 1-7). [6411704] United States: IEEE, Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/DICTA.2012.6411704
Wei, Xingjie ; Li, Chang Tsun ; Hu, Yongjian. / Robust face recognition under varying illumination and occlusion considering structured sparsity. Proceedings of the 2012 International Conference on Digital Image Computing: Techniques and Applications. United States : IEEE, Institute of Electrical and Electronics Engineers, 2012. pp. 1-7
@inproceedings{8762b656e97c4e7da19c992d2c23b3f8,
title = "Robust face recognition under varying illumination and occlusion considering structured sparsity",
abstract = "A large amount of work has been done over the past decades in face recognition (FR). Most of them deal with uncontrolled variations such as changes in illumination, pose, expression and occlusion individually. However, limited work focuses on simultaneously handling multiple variations. In real-world environment, uncontrolled variations usually coexist. FR approaches which are robust to one kind of variation may fail to deal with another. In this paper, we propose an approach considering structured sparsity to deal with the illumination changes and occlusion at the same time. Our approach represents a face image taking into account that the face images usually lie in the structured union of subspaces in a high dimensional feature space. Considering the spatial continuity of the occlusion, we propose a cluster occlusion dictionary for occlusion modelling. In addition, a discriminative feature is embedded in our model to correct the illumination effect. This enables our approach to handle images that lie outside the illumination subspace spanned by the training set. Experimental results on public face databases show that the proposed approach is very robust to large illumination changes and occlusion.",
author = "Xingjie Wei and Li, {Chang Tsun} and Yongjian Hu",
year = "2012",
doi = "10.1109/DICTA.2012.6411704",
language = "English",
pages = "1--7",
booktitle = "Proceedings of the 2012 International Conference on Digital Image Computing: Techniques and Applications",
publisher = "IEEE, Institute of Electrical and Electronics Engineers",
address = "United States",

}

Wei, X, Li, CT & Hu, Y 2012, Robust face recognition under varying illumination and occlusion considering structured sparsity. in Proceedings of the 2012 International Conference on Digital Image Computing: Techniques and Applications., 6411704, IEEE, Institute of Electrical and Electronics Engineers, United States, pp. 1-7, 2012 International Conference on Digital Image Computing: Techniques and Applications (DICTA), Fremantle, Australia, 03/12/12. https://doi.org/10.1109/DICTA.2012.6411704

Robust face recognition under varying illumination and occlusion considering structured sparsity. / Wei, Xingjie; Li, Chang Tsun; Hu, Yongjian.

Proceedings of the 2012 International Conference on Digital Image Computing: Techniques and Applications. United States : IEEE, Institute of Electrical and Electronics Engineers, 2012. p. 1-7 6411704.

Research output: Book chapter/Published conference paperConference paper

TY - GEN

T1 - Robust face recognition under varying illumination and occlusion considering structured sparsity

AU - Wei, Xingjie

AU - Li, Chang Tsun

AU - Hu, Yongjian

PY - 2012

Y1 - 2012

N2 - A large amount of work has been done over the past decades in face recognition (FR). Most of them deal with uncontrolled variations such as changes in illumination, pose, expression and occlusion individually. However, limited work focuses on simultaneously handling multiple variations. In real-world environment, uncontrolled variations usually coexist. FR approaches which are robust to one kind of variation may fail to deal with another. In this paper, we propose an approach considering structured sparsity to deal with the illumination changes and occlusion at the same time. Our approach represents a face image taking into account that the face images usually lie in the structured union of subspaces in a high dimensional feature space. Considering the spatial continuity of the occlusion, we propose a cluster occlusion dictionary for occlusion modelling. In addition, a discriminative feature is embedded in our model to correct the illumination effect. This enables our approach to handle images that lie outside the illumination subspace spanned by the training set. Experimental results on public face databases show that the proposed approach is very robust to large illumination changes and occlusion.

AB - A large amount of work has been done over the past decades in face recognition (FR). Most of them deal with uncontrolled variations such as changes in illumination, pose, expression and occlusion individually. However, limited work focuses on simultaneously handling multiple variations. In real-world environment, uncontrolled variations usually coexist. FR approaches which are robust to one kind of variation may fail to deal with another. In this paper, we propose an approach considering structured sparsity to deal with the illumination changes and occlusion at the same time. Our approach represents a face image taking into account that the face images usually lie in the structured union of subspaces in a high dimensional feature space. Considering the spatial continuity of the occlusion, we propose a cluster occlusion dictionary for occlusion modelling. In addition, a discriminative feature is embedded in our model to correct the illumination effect. This enables our approach to handle images that lie outside the illumination subspace spanned by the training set. Experimental results on public face databases show that the proposed approach is very robust to large illumination changes and occlusion.

UR - http://www.scopus.com/inward/record.url?scp=84874391503&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84874391503&partnerID=8YFLogxK

U2 - 10.1109/DICTA.2012.6411704

DO - 10.1109/DICTA.2012.6411704

M3 - Conference paper

AN - SCOPUS:84874391503

SP - 1

EP - 7

BT - Proceedings of the 2012 International Conference on Digital Image Computing: Techniques and Applications

PB - IEEE, Institute of Electrical and Electronics Engineers

CY - United States

ER -

Wei X, Li CT, Hu Y. Robust face recognition under varying illumination and occlusion considering structured sparsity. In Proceedings of the 2012 International Conference on Digital Image Computing: Techniques and Applications. United States: IEEE, Institute of Electrical and Electronics Engineers. 2012. p. 1-7. 6411704 https://doi.org/10.1109/DICTA.2012.6411704