Mining discriminative components with low-rank and sparsity constraints for face recognition

Qiang Zhang, Baoxin Li

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

15 Citations (Scopus)

Abstract

This paper introduces a novel image decomposition approach for an ensemble of correlated images, using low-rank and sparsity constraints. Each image is decomposed as a combination of three components: one common component, one condition component, which is assumed to be a low-rank matrix, and a sparse residual. For a set of face images of Nsubjects, the decomposition finds N common components, one for each subject, K low-rank components, each capturing a different global condition of the set (e.g., different illumination conditions), and a sparse residual for each input image. Through this decomposition, the proposed approach recovers a clean face image (the common component) for each subject and discovers the conditions (the condition components and the sparse residuals) of the images in the set. The set of N+K images containing only the common and the low-rank components form a compact and discriminative representation for the original images. We design a classifier using only these N+K images. Experiments on commonly-used face data sets demonstrate the effectiveness of the approach for face recognition through comparing with the leading state-of-the-art in the literature. The experiments further show good accuracy in classifying the condition of an input image, suggesting that the components from the proposed decomposition indeed capture physically meaningful features of the input.

Original languageEnglish (US)
Title of host publicationProceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining
Pages1469-1477
Number of pages9
DOIs
StatePublished - 2012
Event18th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2012 - Beijing, China
Duration: Aug 12 2012Aug 16 2012

Other

Other18th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2012
CountryChina
CityBeijing
Period8/12/128/16/12

Fingerprint

Face recognition
Decomposition
Classifiers
Lighting
Experiments

Keywords

  • component decomposition
  • face recognition
  • low-rank matrix
  • sparse matrix
  • subspace learning

ASJC Scopus subject areas

  • Software
  • Information Systems

Cite this

Zhang, Q., & Li, B. (2012). Mining discriminative components with low-rank and sparsity constraints for face recognition. In Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 1469-1477) https://doi.org/10.1145/2339530.2339760

Mining discriminative components with low-rank and sparsity constraints for face recognition. / Zhang, Qiang; Li, Baoxin.

Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2012. p. 1469-1477.

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

Zhang, Q & Li, B 2012, Mining discriminative components with low-rank and sparsity constraints for face recognition. in Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. pp. 1469-1477, 18th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2012, Beijing, China, 8/12/12. https://doi.org/10.1145/2339530.2339760
Zhang Q, Li B. Mining discriminative components with low-rank and sparsity constraints for face recognition. In Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2012. p. 1469-1477 https://doi.org/10.1145/2339530.2339760
Zhang, Qiang ; Li, Baoxin. / Mining discriminative components with low-rank and sparsity constraints for face recognition. Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2012. pp. 1469-1477
@inproceedings{df93732492b64942aaadb049967d9bd8,
title = "Mining discriminative components with low-rank and sparsity constraints for face recognition",
abstract = "This paper introduces a novel image decomposition approach for an ensemble of correlated images, using low-rank and sparsity constraints. Each image is decomposed as a combination of three components: one common component, one condition component, which is assumed to be a low-rank matrix, and a sparse residual. For a set of face images of Nsubjects, the decomposition finds N common components, one for each subject, K low-rank components, each capturing a different global condition of the set (e.g., different illumination conditions), and a sparse residual for each input image. Through this decomposition, the proposed approach recovers a clean face image (the common component) for each subject and discovers the conditions (the condition components and the sparse residuals) of the images in the set. The set of N+K images containing only the common and the low-rank components form a compact and discriminative representation for the original images. We design a classifier using only these N+K images. Experiments on commonly-used face data sets demonstrate the effectiveness of the approach for face recognition through comparing with the leading state-of-the-art in the literature. The experiments further show good accuracy in classifying the condition of an input image, suggesting that the components from the proposed decomposition indeed capture physically meaningful features of the input.",
keywords = "component decomposition, face recognition, low-rank matrix, sparse matrix, subspace learning",
author = "Qiang Zhang and Baoxin Li",
year = "2012",
doi = "10.1145/2339530.2339760",
language = "English (US)",
isbn = "9781450314626",
pages = "1469--1477",
booktitle = "Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining",

}

TY - GEN

T1 - Mining discriminative components with low-rank and sparsity constraints for face recognition

AU - Zhang, Qiang

AU - Li, Baoxin

PY - 2012

Y1 - 2012

N2 - This paper introduces a novel image decomposition approach for an ensemble of correlated images, using low-rank and sparsity constraints. Each image is decomposed as a combination of three components: one common component, one condition component, which is assumed to be a low-rank matrix, and a sparse residual. For a set of face images of Nsubjects, the decomposition finds N common components, one for each subject, K low-rank components, each capturing a different global condition of the set (e.g., different illumination conditions), and a sparse residual for each input image. Through this decomposition, the proposed approach recovers a clean face image (the common component) for each subject and discovers the conditions (the condition components and the sparse residuals) of the images in the set. The set of N+K images containing only the common and the low-rank components form a compact and discriminative representation for the original images. We design a classifier using only these N+K images. Experiments on commonly-used face data sets demonstrate the effectiveness of the approach for face recognition through comparing with the leading state-of-the-art in the literature. The experiments further show good accuracy in classifying the condition of an input image, suggesting that the components from the proposed decomposition indeed capture physically meaningful features of the input.

AB - This paper introduces a novel image decomposition approach for an ensemble of correlated images, using low-rank and sparsity constraints. Each image is decomposed as a combination of three components: one common component, one condition component, which is assumed to be a low-rank matrix, and a sparse residual. For a set of face images of Nsubjects, the decomposition finds N common components, one for each subject, K low-rank components, each capturing a different global condition of the set (e.g., different illumination conditions), and a sparse residual for each input image. Through this decomposition, the proposed approach recovers a clean face image (the common component) for each subject and discovers the conditions (the condition components and the sparse residuals) of the images in the set. The set of N+K images containing only the common and the low-rank components form a compact and discriminative representation for the original images. We design a classifier using only these N+K images. Experiments on commonly-used face data sets demonstrate the effectiveness of the approach for face recognition through comparing with the leading state-of-the-art in the literature. The experiments further show good accuracy in classifying the condition of an input image, suggesting that the components from the proposed decomposition indeed capture physically meaningful features of the input.

KW - component decomposition

KW - face recognition

KW - low-rank matrix

KW - sparse matrix

KW - subspace learning

UR - http://www.scopus.com/inward/record.url?scp=84866020210&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84866020210&partnerID=8YFLogxK

U2 - 10.1145/2339530.2339760

DO - 10.1145/2339530.2339760

M3 - Conference contribution

AN - SCOPUS:84866020210

SN - 9781450314626

SP - 1469

EP - 1477

BT - Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

ER -