Searching Things in Large Sets of Images

Authors
Publication date 2013
Host editors
  • P. van Emde Boas
  • F.C.A. Groen
  • G.F. Italiano
  • J. Nawrocki
  • H. Sack
Book title SOFSEM 2013: Theory and Practice of Computer Science
Book subtitle 39th International Conference on Current Trends in Theory and Practice of Computer Science, Špindlerův Mlýn, Czech Republic, January 26-31, 2013: proceedings
ISBN
  • 9783642358425
ISBN (electronic)
  • 9783642358432
Series Lecture Notes in Computer Science
Pages (from-to) 107
Number of pages 1
Publisher Heidelberg: Springer
Organisations
  • Faculty of Science (FNWI) - Informatics Institute (IVI)
Abstract
In this presentation we discuss the challenges of computer vision in general search through images and videos. The point of departure is the digital content of the images, a set of 50 to 100 examples, and smart features, machine learning and computational tools to find the images best answering a one-word specification. This problem is far from being solved, yet the progress as recorded in the yearly TRECvid competitions for image search engines is impressive. We will discuss the difference between the “where” and “what” in images, discuss the need for invariance, fast computational approaches, and the state of the art in image search engines.
Document type Conference contribution
Language English
Published at https://doi.org/10.1007/978-3-642-35843-2_10
Other links http://www.science.uva.nl/research/publications/2013/SmeuldersTPCS2013
Permalink to this page
Back