Kernel collaborative face recognition

Dong Wang, Huchuan Lu, Ming Hsuan Yang

Research output: Contribution to journalArticle

43 Citations (Scopus)

Abstract

Recent research has demonstrated the effectiveness of linear representation (i.e., sparse representation, group sparse representation and collaborative representation) for face recognition and other vision problems. However, this linear representation assumption does not consider the non-linear relationship of samples and limits the usage of different features with non-linear metrics. In this paper, we present some insights of linear and non-linear representation-based classifiers. First, we present a general formulation known as kernel collaborative representation to encompass several effective representation-based classifiers within a unified framework. Based on this framework, different algorithms can be developed by choosing proper kernel functions, regularization terms, and additional constraints. Second, within the proposed framework we develop a simple yet effective algorithm with squared ℓ2-regularization and apply it to face recognition with local binary patterns as well as the Hamming kernel. We conduct numerous experiments on the extended Yale B, AR, Multi-PIE, PloyU NIR, PloyU HS, EURECOM Kinect and FERET face databases. Experimental results demonstrate that our algorithm achieves favorable performance in terms of accuracy and speed, especially for the face recognition problems with small training datasets and heavy occlusion. In addition, we attempt to combine different kernel functions by using different weights in an additive manner. The experimental results show that the proposed combination scheme provides some additional improvement in terms of accuracy.

Original languageEnglish
Pages (from-to)3025-3037
Number of pages13
JournalPattern Recognition
Volume48
Issue number10
DOIs
Publication statusPublished - 2015 Oct 1

Fingerprint

Face recognition
Classifiers
Experiments

All Science Journal Classification (ASJC) codes

  • Software
  • Signal Processing
  • Computer Vision and Pattern Recognition
  • Artificial Intelligence

Cite this

Wang, Dong ; Lu, Huchuan ; Yang, Ming Hsuan. / Kernel collaborative face recognition. In: Pattern Recognition. 2015 ; Vol. 48, No. 10. pp. 3025-3037.
@article{dc9869b6baaf47d3b4873a82070c6c6e,
title = "Kernel collaborative face recognition",
abstract = "Recent research has demonstrated the effectiveness of linear representation (i.e., sparse representation, group sparse representation and collaborative representation) for face recognition and other vision problems. However, this linear representation assumption does not consider the non-linear relationship of samples and limits the usage of different features with non-linear metrics. In this paper, we present some insights of linear and non-linear representation-based classifiers. First, we present a general formulation known as kernel collaborative representation to encompass several effective representation-based classifiers within a unified framework. Based on this framework, different algorithms can be developed by choosing proper kernel functions, regularization terms, and additional constraints. Second, within the proposed framework we develop a simple yet effective algorithm with squared ℓ2-regularization and apply it to face recognition with local binary patterns as well as the Hamming kernel. We conduct numerous experiments on the extended Yale B, AR, Multi-PIE, PloyU NIR, PloyU HS, EURECOM Kinect and FERET face databases. Experimental results demonstrate that our algorithm achieves favorable performance in terms of accuracy and speed, especially for the face recognition problems with small training datasets and heavy occlusion. In addition, we attempt to combine different kernel functions by using different weights in an additive manner. The experimental results show that the proposed combination scheme provides some additional improvement in terms of accuracy.",
author = "Dong Wang and Huchuan Lu and Yang, {Ming Hsuan}",
year = "2015",
month = "10",
day = "1",
doi = "10.1016/j.patcog.2015.01.012",
language = "English",
volume = "48",
pages = "3025--3037",
journal = "Pattern Recognition",
issn = "0031-3203",
publisher = "Elsevier Limited",
number = "10",

}

Kernel collaborative face recognition. / Wang, Dong; Lu, Huchuan; Yang, Ming Hsuan.

In: Pattern Recognition, Vol. 48, No. 10, 01.10.2015, p. 3025-3037.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Kernel collaborative face recognition

AU - Wang, Dong

AU - Lu, Huchuan

AU - Yang, Ming Hsuan

PY - 2015/10/1

Y1 - 2015/10/1

N2 - Recent research has demonstrated the effectiveness of linear representation (i.e., sparse representation, group sparse representation and collaborative representation) for face recognition and other vision problems. However, this linear representation assumption does not consider the non-linear relationship of samples and limits the usage of different features with non-linear metrics. In this paper, we present some insights of linear and non-linear representation-based classifiers. First, we present a general formulation known as kernel collaborative representation to encompass several effective representation-based classifiers within a unified framework. Based on this framework, different algorithms can be developed by choosing proper kernel functions, regularization terms, and additional constraints. Second, within the proposed framework we develop a simple yet effective algorithm with squared ℓ2-regularization and apply it to face recognition with local binary patterns as well as the Hamming kernel. We conduct numerous experiments on the extended Yale B, AR, Multi-PIE, PloyU NIR, PloyU HS, EURECOM Kinect and FERET face databases. Experimental results demonstrate that our algorithm achieves favorable performance in terms of accuracy and speed, especially for the face recognition problems with small training datasets and heavy occlusion. In addition, we attempt to combine different kernel functions by using different weights in an additive manner. The experimental results show that the proposed combination scheme provides some additional improvement in terms of accuracy.

AB - Recent research has demonstrated the effectiveness of linear representation (i.e., sparse representation, group sparse representation and collaborative representation) for face recognition and other vision problems. However, this linear representation assumption does not consider the non-linear relationship of samples and limits the usage of different features with non-linear metrics. In this paper, we present some insights of linear and non-linear representation-based classifiers. First, we present a general formulation known as kernel collaborative representation to encompass several effective representation-based classifiers within a unified framework. Based on this framework, different algorithms can be developed by choosing proper kernel functions, regularization terms, and additional constraints. Second, within the proposed framework we develop a simple yet effective algorithm with squared ℓ2-regularization and apply it to face recognition with local binary patterns as well as the Hamming kernel. We conduct numerous experiments on the extended Yale B, AR, Multi-PIE, PloyU NIR, PloyU HS, EURECOM Kinect and FERET face databases. Experimental results demonstrate that our algorithm achieves favorable performance in terms of accuracy and speed, especially for the face recognition problems with small training datasets and heavy occlusion. In addition, we attempt to combine different kernel functions by using different weights in an additive manner. The experimental results show that the proposed combination scheme provides some additional improvement in terms of accuracy.

UR - http://www.scopus.com/inward/record.url?scp=84931566105&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84931566105&partnerID=8YFLogxK

U2 - 10.1016/j.patcog.2015.01.012

DO - 10.1016/j.patcog.2015.01.012

M3 - Article

AN - SCOPUS:84931566105

VL - 48

SP - 3025

EP - 3037

JO - Pattern Recognition

JF - Pattern Recognition

SN - 0031-3203

IS - 10

ER -