Skip to main navigation Skip to search Skip to main content

Maximum constrained sparse coding for image representation

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Sparse coding exhibits good performance in many computer vision applications by finding bases which capture highlevel semantics of the data and learning sparse coefficients in terms of the bases. However, due to the fact that bases are non-orthogonal, sparse coding can hardly preserve the samplesâ™ similarity, which is important for discrimination. In this paper, a new image representing method called maximum constrained sparse coding (MCSC) is proposed. Sparse representation with more active coefficients means more similarity information, and the infinite norm is added to the solution for this purpose. We solve the optimizer by constraining the codesâ™ maximum and releasing the residual to other dictionary atoms. Experimental results on image clustering show that our method can preserve the similarity of adjacent samples and maintain the sparsity of code simultaneously.

Original languageEnglish
Title of host publicationMIPPR 2015
Subtitle of host publicationPattern Recognition and Computer Vision
EditorsJianguo Liu, Tianxu Zhang
PublisherSPIE
ISBN (Electronic)9781510600546
DOIs
StatePublished - 2015
Event9th International Symposium on Multispectral Image Processing and Pattern Recognition, MIPPR 2015 - Enshi, Hubei, China
Duration: 31 Oct 20151 Nov 2015

Publication series

NameProceedings of SPIE - The International Society for Optical Engineering
Volume9813
ISSN (Print)0277-786X
ISSN (Electronic)1996-756X

Conference

Conference9th International Symposium on Multispectral Image Processing and Pattern Recognition, MIPPR 2015
Country/TerritoryChina
CityEnshi, Hubei
Period31/10/151/11/15

Keywords

  • Sparse coding
  • image representation
  • infinite norm
  • maximum constraint

Fingerprint

Dive into the research topics of 'Maximum constrained sparse coding for image representation'. Together they form a unique fingerprint.

Cite this