Vis enkel innførsel

dc.contributor.authorLiu, Wei-Xue
dc.contributor.authorHou, Jian
dc.contributor.authorKarimi, Hamid Reza
dc.date.accessioned2014-12-17T08:58:10Z
dc.date.available2014-12-17T08:58:10Z
dc.date.issued2014
dc.identifier.citationLiu, W.-X., Hou, J., & Karimi, H. R. (2014). Research on vocabulary sizes and codebook universality. Abstract and Applied Analysis, 2014, 1-7. doi: 10.1155/2014/697245nb_NO
dc.identifier.issn1687-0409
dc.identifier.urihttp://hdl.handle.net/11250/227586
dc.descriptionPublished version of an article in the journal: Abstract and Applied Analysis. Also available from the publisher at: http://dx.doi.org/10.1155/2014/697245 Open Accessnb_NO
dc.description.abstractCodebook is an effective image representation method. By clustering in local image descriptors, a codebook is shown to be a distinctive image feature and widely applied in object classification. In almost all existing works on codebooks, the building of the visual vocabulary follows a basic routine, that is, extracting local image descriptors and clustering with a user-designated number of clusters. The problem with this routine lies in that building a codebook for each single dataset is not efficient. In order to deal with this problem, we investigate the influence of vocabulary sizes on classification performance and vocabulary universality with the kNN classifier. Experimental results indicate that, under the condition that the vocabulary size is large enough, the vocabularies built from different datasets are exchangeable and universal.nb_NO
dc.language.isoengnb_NO
dc.publisherHindawi Publishing Corporationnb_NO
dc.titleResearch on vocabulary sizes and codebook universalitynb_NO
dc.typeJournal articlenb_NO
dc.typePeer reviewednb_NO
dc.subject.nsiVDP::Technology: 500::Information and communication technology: 550nb_NO
dc.source.pagenumber1-7nb_NO
dc.source.journalAbstract and Applied Analysisnb_NO
dc.identifier.doi10.1155/2014/697245


Tilhørende fil(er)

Thumbnail

Denne innførselen finnes i følgende samling(er)

Vis enkel innførsel