Image Clustering Using Multimodal Keywords

  • Rajeev Agrawal
  • William Grosky
  • Farshad Fotouhi
Conference paper

DOI: 10.1007/11930334_9

Part of the Lecture Notes in Computer Science book series (LNCS, volume 4306)
Cite this paper as:
Agrawal R., Grosky W., Fotouhi F. (2006) Image Clustering Using Multimodal Keywords. In: Avrithis Y., Kompatsiaris Y., Staab S., O’Connor N.E. (eds) Semantic Multimedia. SAMT 2006. Lecture Notes in Computer Science, vol 4306. Springer, Berlin, Heidelberg

Abstract

Extending our previous work on visual keywords, we use the concept of template-based visual keywords using MPEG-7 color descriptors. MPEG-7, also called the Multimedia Content Description Interface, has been a standard for many years. These color descriptors have the ability to characterize perceptual color similarity and need relatively low complexity operations to extract them, besides being scalable and interoperable. We then demonstrate the power of these visual keywords for image clustering, when used in tandem with textual keyword annotations, in the context of latent semantic analysis, a popular technique in classical information retrieval which has been used to reveal the underlying semantic structure of document collections.

Keywords

MPEG-7 visual keywords textual keywords latent semantic analysis singular value decomposition adjusted rand index 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Rajeev Agrawal
    • 1
    • 2
  • William Grosky
    • 3
  • Farshad Fotouhi
    • 2
  1. 1.Kettering UniversityFlintUSA
  2. 2.Wayne State UniversityDetroitUSA
  3. 3.The University of Michigan – DearbornDearbornUSA

Personalised recommendations