Gates Building, Room 100, 3:15-4:30
Friday, 12 Feb, 1999
Search engines are useful because they allow the user to find information of interest from the World-Wide Web. However, most of the popular search engines today are textual: they do not allow the user to find images from the Web.
In this talk we explain AMORE, a Web search engine that allows the user to retrieve images from the Web by specifying relevant keywords or by requesting an image similar to a given image. Text and image search can also be combined.
In the indexing phases, the system extracts the objects from the image and assigns relevant keywords to the images based on the HTML page containing the image. Using these metadata, users can retrieve images which are visually and/or semantically similar to the user-specified image.
We explain the architecture of the system, and mechanisms for the indexing and matching phases. We will also explain our experience in using the AMORE site and the Getty Art Museum site application.