Face recognition with image sets using manifold density divergence

Ognjen Arandjelovic; Gregory Shakhnarovich; John Fisher; Roberto Cipolla; Trevor Darrell

doi:10.1109/CVPR.2005.151

Face recognition with image sets using manifold density divergence

Ognjen Arandjelovic^*
, Gregory Shakhnarovich
, John Fisher
, Roberto Cipolla
, Trevor Darrell

^*Corresponding author for this work

School of Computer Science

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

Abstract

In many automatic face recognition applications, a set of a person's face images is available rather than a single image. In this paper, we describe a novel method for face recognition using image sets. We propose a flexible, semiparametric model for learning probability densities confined to highly non-linear but intrinsically low-dimensional manifolds. The model leads to a statistical formulation of the recognition problem in terms of minimizing the divergence between densities estimated on these manifolds. The proposed method is evaluated on a large data set, acquired in realistic imaging conditions with severe illumination variation. Our algorithm is shown to match the best and outperform other state-of-the-art algorithms in the literature, achieving 94% recognition rate on average.

Original language	English
Title of host publication	Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
Editors	C. Schmid, S. Soatto, C. Tomasi
Pages	581-588
Number of pages	8
Volume	1
DOIs	https://doi.org/10.1109/CVPR.2005.151
Publication status	Published - 2005
Event	2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2005 - San Diego, CA, United States Duration: 20 Jun 2005 → 25 Jun 2005

Conference

Conference	2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2005
Country/Territory	United States
City	San Diego, CA
Period	20/06/05 → 25/06/05

Access to Document

10.1109/CVPR.2005.151

Handle.net

Persistent link

Cite this

@inproceedings{bbfded0e95a9412d909cc7e33bdd75c5,

title = "Face recognition with image sets using manifold density divergence",

abstract = "In many automatic face recognition applications, a set of a person's face images is available rather than a single image. In this paper, we describe a novel method for face recognition using image sets. We propose a flexible, semiparametric model for learning probability densities confined to highly non-linear but intrinsically low-dimensional manifolds. The model leads to a statistical formulation of the recognition problem in terms of minimizing the divergence between densities estimated on these manifolds. The proposed method is evaluated on a large data set, acquired in realistic imaging conditions with severe illumination variation. Our algorithm is shown to match the best and outperform other state-of-the-art algorithms in the literature, achieving 94\% recognition rate on average.",

author = "Ognjen Arandjelovic and Gregory Shakhnarovich and John Fisher and Roberto Cipolla and Trevor Darrell",

year = "2005",

doi = "10.1109/CVPR.2005.151",

language = "English",

isbn = "0769523722",

volume = "1",

pages = "581--588",

editor = "C. Schmid and S. Soatto and C. Tomasi",

booktitle = "Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition",

note = "2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2005 ; Conference date: 20-06-2005 Through 25-06-2005",

}

Arandjelovic, O, Shakhnarovich, G, Fisher, J, Cipolla, R & Darrell, T 2005, Face recognition with image sets using manifold density divergence. in C Schmid, S Soatto & C Tomasi (eds), Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. vol. 1, pp. 581-588, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2005, San Diego, CA, United States, 20/06/05. https://doi.org/10.1109/CVPR.2005.151

Face recognition with image sets using manifold density divergence. / Arandjelovic, Ognjen; Shakhnarovich, Gregory; Fisher, John et al.
Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. ed. / C. Schmid; S. Soatto; C. Tomasi. Vol. 1 2005. p. 581-588.

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

TY - GEN

T1 - Face recognition with image sets using manifold density divergence

AU - Arandjelovic, Ognjen

AU - Shakhnarovich, Gregory

AU - Fisher, John

AU - Cipolla, Roberto

AU - Darrell, Trevor

PY - 2005

Y1 - 2005

N2 - In many automatic face recognition applications, a set of a person's face images is available rather than a single image. In this paper, we describe a novel method for face recognition using image sets. We propose a flexible, semiparametric model for learning probability densities confined to highly non-linear but intrinsically low-dimensional manifolds. The model leads to a statistical formulation of the recognition problem in terms of minimizing the divergence between densities estimated on these manifolds. The proposed method is evaluated on a large data set, acquired in realistic imaging conditions with severe illumination variation. Our algorithm is shown to match the best and outperform other state-of-the-art algorithms in the literature, achieving 94% recognition rate on average.

AB - In many automatic face recognition applications, a set of a person's face images is available rather than a single image. In this paper, we describe a novel method for face recognition using image sets. We propose a flexible, semiparametric model for learning probability densities confined to highly non-linear but intrinsically low-dimensional manifolds. The model leads to a statistical formulation of the recognition problem in terms of minimizing the divergence between densities estimated on these manifolds. The proposed method is evaluated on a large data set, acquired in realistic imaging conditions with severe illumination variation. Our algorithm is shown to match the best and outperform other state-of-the-art algorithms in the literature, achieving 94% recognition rate on average.

UR - https://www.scopus.com/pages/publications/24644491332

U2 - 10.1109/CVPR.2005.151

DO - 10.1109/CVPR.2005.151

M3 - Conference contribution

AN - SCOPUS:24644491332

SN - 0769523722

VL - 1

SP - 581

EP - 588

BT - Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition

A2 - Schmid, C.

A2 - Soatto, S.

A2 - Tomasi, C.

T2 - 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2005

Y2 - 20 June 2005 through 25 June 2005

ER -

Face recognition with image sets using manifold density divergence

Abstract

Conference

Access to Document

Handle.net

Other files and links

Fingerprint

Cite this