Low-rank sparse coding for image classification

Tianzhu Zhang, Bernard Ghanem, Si Liu, Changsheng Xu, Narendra Ahuja

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

Abstract

In this paper, we propose a low-rank sparse coding (LRSC) method that exploits local structure information among features in an image for the purpose of image-level classification. LRSC represents densely sampled SIFT descriptors, in a spatial neighborhood, collectively as low-rank, sparse linear combinations of code words. As such, it casts the feature coding problem as a low-rank matrix learning problem, which is different from previous methods that encode features independently. This LRSC has a number of attractive properties. (1) It encourages sparsity in feature codes, locality in codebook construction, and low-rankness for spatial consistency. (2) LRSC encodes local features jointly by considering their low-rank structure information, and is computationally attractive. We evaluate the LRSC by comparing its performance on a set of challenging benchmarks with that of 7 popular coding and other state-of-the-art methods. Our experiments show that by representing local features jointly, LRSC not only outperforms the state-of-the-art in classification accuracy but also improves the time complexity of methods that use a similar sparse linear representation model for feature coding.

Original languageEnglish (US)
Title of host publicationProceedings - 2013 IEEE International Conference on Computer Vision, ICCV 2013
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages281-288
Number of pages8
ISBN (Print)9781479928392
DOIs
StatePublished - Jan 1 2013
Event2013 14th IEEE International Conference on Computer Vision, ICCV 2013 - Sydney, NSW, Australia
Duration: Dec 1 2013Dec 8 2013

Publication series

NameProceedings of the IEEE International Conference on Computer Vision

Other

Other2013 14th IEEE International Conference on Computer Vision, ICCV 2013
CountryAustralia
CitySydney, NSW
Period12/1/1312/8/13

Fingerprint

Image classification
Experiments

Keywords

  • bow
  • image classification
  • low-rank

ASJC Scopus subject areas

  • Software
  • Computer Vision and Pattern Recognition

Cite this

Zhang, T., Ghanem, B., Liu, S., Xu, C., & Ahuja, N. (2013). Low-rank sparse coding for image classification. In Proceedings - 2013 IEEE International Conference on Computer Vision, ICCV 2013 (pp. 281-288). [6751144] (Proceedings of the IEEE International Conference on Computer Vision). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICCV.2013.42

Low-rank sparse coding for image classification. / Zhang, Tianzhu; Ghanem, Bernard; Liu, Si; Xu, Changsheng; Ahuja, Narendra.

Proceedings - 2013 IEEE International Conference on Computer Vision, ICCV 2013. Institute of Electrical and Electronics Engineers Inc., 2013. p. 281-288 6751144 (Proceedings of the IEEE International Conference on Computer Vision).

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

Zhang, T, Ghanem, B, Liu, S, Xu, C & Ahuja, N 2013, Low-rank sparse coding for image classification. in Proceedings - 2013 IEEE International Conference on Computer Vision, ICCV 2013., 6751144, Proceedings of the IEEE International Conference on Computer Vision, Institute of Electrical and Electronics Engineers Inc., pp. 281-288, 2013 14th IEEE International Conference on Computer Vision, ICCV 2013, Sydney, NSW, Australia, 12/1/13. https://doi.org/10.1109/ICCV.2013.42
Zhang T, Ghanem B, Liu S, Xu C, Ahuja N. Low-rank sparse coding for image classification. In Proceedings - 2013 IEEE International Conference on Computer Vision, ICCV 2013. Institute of Electrical and Electronics Engineers Inc. 2013. p. 281-288. 6751144. (Proceedings of the IEEE International Conference on Computer Vision). https://doi.org/10.1109/ICCV.2013.42
Zhang, Tianzhu ; Ghanem, Bernard ; Liu, Si ; Xu, Changsheng ; Ahuja, Narendra. / Low-rank sparse coding for image classification. Proceedings - 2013 IEEE International Conference on Computer Vision, ICCV 2013. Institute of Electrical and Electronics Engineers Inc., 2013. pp. 281-288 (Proceedings of the IEEE International Conference on Computer Vision).
@inproceedings{232ea98e28c44d019ae766602dd5a911,
title = "Low-rank sparse coding for image classification",
abstract = "In this paper, we propose a low-rank sparse coding (LRSC) method that exploits local structure information among features in an image for the purpose of image-level classification. LRSC represents densely sampled SIFT descriptors, in a spatial neighborhood, collectively as low-rank, sparse linear combinations of code words. As such, it casts the feature coding problem as a low-rank matrix learning problem, which is different from previous methods that encode features independently. This LRSC has a number of attractive properties. (1) It encourages sparsity in feature codes, locality in codebook construction, and low-rankness for spatial consistency. (2) LRSC encodes local features jointly by considering their low-rank structure information, and is computationally attractive. We evaluate the LRSC by comparing its performance on a set of challenging benchmarks with that of 7 popular coding and other state-of-the-art methods. Our experiments show that by representing local features jointly, LRSC not only outperforms the state-of-the-art in classification accuracy but also improves the time complexity of methods that use a similar sparse linear representation model for feature coding.",
keywords = "bow, image classification, low-rank",
author = "Tianzhu Zhang and Bernard Ghanem and Si Liu and Changsheng Xu and Narendra Ahuja",
year = "2013",
month = "1",
day = "1",
doi = "10.1109/ICCV.2013.42",
language = "English (US)",
isbn = "9781479928392",
series = "Proceedings of the IEEE International Conference on Computer Vision",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "281--288",
booktitle = "Proceedings - 2013 IEEE International Conference on Computer Vision, ICCV 2013",
address = "United States",

}

TY - GEN

T1 - Low-rank sparse coding for image classification

AU - Zhang, Tianzhu

AU - Ghanem, Bernard

AU - Liu, Si

AU - Xu, Changsheng

AU - Ahuja, Narendra

PY - 2013/1/1

Y1 - 2013/1/1

N2 - In this paper, we propose a low-rank sparse coding (LRSC) method that exploits local structure information among features in an image for the purpose of image-level classification. LRSC represents densely sampled SIFT descriptors, in a spatial neighborhood, collectively as low-rank, sparse linear combinations of code words. As such, it casts the feature coding problem as a low-rank matrix learning problem, which is different from previous methods that encode features independently. This LRSC has a number of attractive properties. (1) It encourages sparsity in feature codes, locality in codebook construction, and low-rankness for spatial consistency. (2) LRSC encodes local features jointly by considering their low-rank structure information, and is computationally attractive. We evaluate the LRSC by comparing its performance on a set of challenging benchmarks with that of 7 popular coding and other state-of-the-art methods. Our experiments show that by representing local features jointly, LRSC not only outperforms the state-of-the-art in classification accuracy but also improves the time complexity of methods that use a similar sparse linear representation model for feature coding.

AB - In this paper, we propose a low-rank sparse coding (LRSC) method that exploits local structure information among features in an image for the purpose of image-level classification. LRSC represents densely sampled SIFT descriptors, in a spatial neighborhood, collectively as low-rank, sparse linear combinations of code words. As such, it casts the feature coding problem as a low-rank matrix learning problem, which is different from previous methods that encode features independently. This LRSC has a number of attractive properties. (1) It encourages sparsity in feature codes, locality in codebook construction, and low-rankness for spatial consistency. (2) LRSC encodes local features jointly by considering their low-rank structure information, and is computationally attractive. We evaluate the LRSC by comparing its performance on a set of challenging benchmarks with that of 7 popular coding and other state-of-the-art methods. Our experiments show that by representing local features jointly, LRSC not only outperforms the state-of-the-art in classification accuracy but also improves the time complexity of methods that use a similar sparse linear representation model for feature coding.

KW - bow

KW - image classification

KW - low-rank

UR - http://www.scopus.com/inward/record.url?scp=84898773327&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84898773327&partnerID=8YFLogxK

U2 - 10.1109/ICCV.2013.42

DO - 10.1109/ICCV.2013.42

M3 - Conference contribution

AN - SCOPUS:84898773327

SN - 9781479928392

T3 - Proceedings of the IEEE International Conference on Computer Vision

SP - 281

EP - 288

BT - Proceedings - 2013 IEEE International Conference on Computer Vision, ICCV 2013

PB - Institute of Electrical and Electronics Engineers Inc.

ER -