Efficient classification with sparsity augmented collaborative representation

Naveed Akhtar, Faisal Shafait, Ajmal Mian

Research output: Contribution to journalArticle

  • 4 Citations

Abstract

Many classification approaches first represent a test sample using the training samples of all the classes. This collaborative representation is then used to label the test sample. It is a common belief that sparseness of the representation is the key to success for this classification scheme. However, more recently, it has been claimed that it is the collaboration and not the sparseness that makes the scheme effective. This claim is attractive as it allows to relinquish the computationally expensive sparsity constraint over the representation. In this paper, we first extend the analysis supporting this claim and then show that sparseness explicitly contributes to improved classification, hence it should not be completely ignored for computational gains. Inspired by this result, we augment a dense collaborative representation with a sparse representation and propose an efficient classification method that capitalizes on the resulting representation. The augmented representation and the classification method work together meticulously to achieve higher accuracy and lower computational time compared to state-of-the-art collaborative representation based classification approaches. Experiments on benchmark face, object, action and scene databases show the efficacy of our approach.

LanguageEnglish
Pages136-145
Number of pages10
JournalPattern Recognition
Volume65
DOIs
StatePublished - 1 May 2017

Fingerprint

Labels
Experiments

Cite this

@article{2d2e720773674fdb9cbbba676252db55,
title = "Efficient classification with sparsity augmented collaborative representation",
abstract = "Many classification approaches first represent a test sample using the training samples of all the classes. This collaborative representation is then used to label the test sample. It is a common belief that sparseness of the representation is the key to success for this classification scheme. However, more recently, it has been claimed that it is the collaboration and not the sparseness that makes the scheme effective. This claim is attractive as it allows to relinquish the computationally expensive sparsity constraint over the representation. In this paper, we first extend the analysis supporting this claim and then show that sparseness explicitly contributes to improved classification, hence it should not be completely ignored for computational gains. Inspired by this result, we augment a dense collaborative representation with a sparse representation and propose an efficient classification method that capitalizes on the resulting representation. The augmented representation and the classification method work together meticulously to achieve higher accuracy and lower computational time compared to state-of-the-art collaborative representation based classification approaches. Experiments on benchmark face, object, action and scene databases show the efficacy of our approach.",
keywords = "Collaborative representation, Multi-class classification, Sparse representation",
author = "Naveed Akhtar and Faisal Shafait and Ajmal Mian",
year = "2017",
month = "5",
day = "1",
doi = "10.1016/j.patcog.2016.12.017",
language = "English",
volume = "65",
pages = "136--145",
journal = "Pattern Recognition",
issn = "0031-3203",
publisher = "Elsevier",

}

Efficient classification with sparsity augmented collaborative representation. / Akhtar, Naveed; Shafait, Faisal; Mian, Ajmal.

In: Pattern Recognition, Vol. 65, 01.05.2017, p. 136-145.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Efficient classification with sparsity augmented collaborative representation

AU - Akhtar,Naveed

AU - Shafait,Faisal

AU - Mian,Ajmal

PY - 2017/5/1

Y1 - 2017/5/1

N2 - Many classification approaches first represent a test sample using the training samples of all the classes. This collaborative representation is then used to label the test sample. It is a common belief that sparseness of the representation is the key to success for this classification scheme. However, more recently, it has been claimed that it is the collaboration and not the sparseness that makes the scheme effective. This claim is attractive as it allows to relinquish the computationally expensive sparsity constraint over the representation. In this paper, we first extend the analysis supporting this claim and then show that sparseness explicitly contributes to improved classification, hence it should not be completely ignored for computational gains. Inspired by this result, we augment a dense collaborative representation with a sparse representation and propose an efficient classification method that capitalizes on the resulting representation. The augmented representation and the classification method work together meticulously to achieve higher accuracy and lower computational time compared to state-of-the-art collaborative representation based classification approaches. Experiments on benchmark face, object, action and scene databases show the efficacy of our approach.

AB - Many classification approaches first represent a test sample using the training samples of all the classes. This collaborative representation is then used to label the test sample. It is a common belief that sparseness of the representation is the key to success for this classification scheme. However, more recently, it has been claimed that it is the collaboration and not the sparseness that makes the scheme effective. This claim is attractive as it allows to relinquish the computationally expensive sparsity constraint over the representation. In this paper, we first extend the analysis supporting this claim and then show that sparseness explicitly contributes to improved classification, hence it should not be completely ignored for computational gains. Inspired by this result, we augment a dense collaborative representation with a sparse representation and propose an efficient classification method that capitalizes on the resulting representation. The augmented representation and the classification method work together meticulously to achieve higher accuracy and lower computational time compared to state-of-the-art collaborative representation based classification approaches. Experiments on benchmark face, object, action and scene databases show the efficacy of our approach.

KW - Collaborative representation

KW - Multi-class classification

KW - Sparse representation

UR - http://www.scopus.com/inward/record.url?scp=85010635550&partnerID=8YFLogxK

U2 - 10.1016/j.patcog.2016.12.017

DO - 10.1016/j.patcog.2016.12.017

M3 - Article

VL - 65

SP - 136

EP - 145

JO - Pattern Recognition

T2 - Pattern Recognition

JF - Pattern Recognition

SN - 0031-3203

ER -