A method for evaluating the performance of content-based image retrieval systems based on subjectively determined similarity between images

John A. Black, Gamal Fahmy, Sethuraman Panchanathan

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

6 Scopus citations

Abstract

In recent years multimedia researchers have attempted to design content-based image retrieval systems. However, despite the development of these systems, the term "content" has still remained rather ill defined, and this has made the evaluation of such systems problematic. This paper proposes a method for the creation of a reference image set in which the similarity of each image pair is estimated by two independent methods – by the subjective evaluation of human observers, and by the use of “visual content words” as basis vectors that allow the multidimensional content of each image to be represented with a content vector. The similarity measure computed with these content vectors is shown to correlate with the subjective judgment of human observers, and thus provides both a more objective method for evaluating and expressing image content, and a possible path to automating the process of content-based indexing in the future.

Original languageEnglish (US)
Title of host publicationImage and Video Retrieval - International Conference, CIVR 2002, Proceedings
EditorsMichael S. Lew, Nicu Sebe, John P. Eakins
PublisherSpringer Verlag
Pages356-366
Number of pages11
ISBN (Electronic)9783540438991
DOIs
StatePublished - 2002
EventInternational Conference on Image and Video Retrieval, CIVR 2002 - London, United Kingdom
Duration: Jul 18 2002Jul 19 2002

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume2383
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

OtherInternational Conference on Image and Video Retrieval, CIVR 2002
Country/TerritoryUnited Kingdom
CityLondon
Period7/18/027/19/02

ASJC Scopus subject areas

  • Theoretical Computer Science
  • General Computer Science

Fingerprint

Dive into the research topics of 'A method for evaluating the performance of content-based image retrieval systems based on subjectively determined similarity between images'. Together they form a unique fingerprint.

Cite this