Exploitation du contenu visuel pour améliorer la recherche textuelle d’images en lignes

Download
  1. (PDF, 521 KB)
  2. Get@NRC: Exploitation du contenu visuel pour améliorer la recherche textuelle d’images en lignes (Opens in a new window)
AuthorSearch for: ; Search for: ; Search for: ; Search for: ; Search for: ; Search for:
TypeArticle
Journal titleDocument numérique
Volume13
Issue1
Pages187209; # of pages: 23
SubjectImages retrieval; Visual concepts; Fuzzy decision trees; Diversity; ImageClef; Concepts visuels; Arbre de décision flou; Diversité; ImageClef
AbstractWeb image search engines tend use the text information associated to the images to find the relevant ones. The visual content is rarely used in the "on-line" phase. We propose a complete processing chain exhibiting two efficient ways to use the visual content on the fly. The first method focuses on improving the retrieval precision by filtering text based results, using visual concepts identified in the (text) query. The visual concepts were previously learned (offline) using Forest of Fuzzy Decision Trees. There is a clear score improvement when concepts explicitly mentioned in the query are used. The second method focuses on the diversity of the results. We propose to partition the visual space and show that it is actually effective.
Publication date
LanguageFrench
AffiliationNRC Institute for Information Technology; National Research Council Canada
Peer reviewedYes
NPARC number16885333
Export citationExport as RIS
Report a correctionReport a correction
Record identifier7c6481d0-bd9d-42a5-a1c6-6c4b11f14baf
Record created2011-02-22
Record modified2016-05-09
Bookmark and share
  • Share this page with Facebook (Opens in a new window)
  • Share this page with Twitter (Opens in a new window)
  • Share this page with Google+ (Opens in a new window)
  • Share this page with Delicious (Opens in a new window)