New Paradigms in Multimedia Management and Search

Stefan Rueger

Department of Computing, Imperial College London

Mon Mar 22 15:10:00 NZST 2004 in Room 031, MSCS

Abstract

This is about content-based multimedia retrieval, eg, search of still images similar to given ones; finding a news story broadcast over TV by providing visual examples; finding a music piece by humming it etc. Our retrieval approach is not based on manual annotation but on /automated/ processing and feature extraction.

Some of the challenges in this approach are given by (i) the semantic gap between what computers can index and high-level human concepts and (ii) polysemy which is inherent in visual material and the corresponding wide range of information need by the user. We try to overcome these challenges with interactive retrieval methods, eg, the use of relevance feedback and browsing, thus putting the user at centre stage. Behind the scene we deploy learning algorithms that adapt themselves to the user and their information need.

We argue that these methods, when integrated into digital libraries, will not only enhance their searching and browsing capabilities but also give access through unconventional query methods such as sketching, similarity browsing and providing examples of what is relevant. The talk will cover current work of our multimedia information retrieval team at Imperial College London:

• News Search Engine and Summarisation
• Search, Relevance Feedback and Browsing in Image and Video Databases
• Information Navigation within a set of Documents