Towards more precise social image-tag alignment

Ning Zhou, Jinye Peng, Xiaoyi Feng, Jianping Fan

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

2 Scopus citations

Abstract

Large-scale user contributed images with tags are increasingly available on the Internet. However, the uncertainty of the relatedness between the images and the tags prohibit them from being precisely accessible to the public and being leveraged for computer vision tasks. In this paper, a novel algorithm is proposed to better align the images with the social tags. First, image clustering is performed to group the images into a set of image clusters based on their visual similarity contexts. By clustering images into different groups, the uncertainty of the relatedness between images and tags can be significantly reduced. Second, random walk is adopted to re-rank the tags based on a cross-modal tag correlation network which harnesses both image visual similarity contexts and tag co-occurrences. We have evaluated the proposed algorithm on a large-scale Flickr data set and achieved very positive results.

Original languageEnglish
Title of host publicationAdvances in Multimedia Modeling - 17th International Multimedia Modeling Conference, MMM 2011, Proceedings
Pages46-56
Number of pages11
EditionPART 2
DOIs
StatePublished - 2011
Event17th Multimedia Modeling Conference, MMM 2011 - Taipei, Taiwan, Province of China
Duration: 5 Jan 20117 Jan 2011

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
NumberPART 2
Volume6524 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference17th Multimedia Modeling Conference, MMM 2011
Country/TerritoryTaiwan, Province of China
CityTaipei
Period5/01/117/01/11

Keywords

  • Image-tag alignment
  • relevance re-ranking
  • tag correlation network

Fingerprint

Dive into the research topics of 'Towards more precise social image-tag alignment'. Together they form a unique fingerprint.

Cite this