VERY LARGE COLLECTIONS OF IMAGES are now common. Indexing and searching such collections using indexing languages is difficult. Computer vision offers a variety of techniques for searching for pictures in large collections. Appearance methods compare images based on the overall content of the image using such criteria as similarity of color histograms, texture histograms, spatial layout, and filtered representations. Finding methods concentrate on matching subparts of images, defined in a variety of ways, in the hope of finding particular objects. These ideas are illustrated with a variety of examples from the current literature.
|Original language||English (US)|
|Number of pages||30|
|State||Published - Sep 1 1999|
ASJC Scopus subject areas
- Library and Information Sciences