About

The overall aim of the project was to facilitate visual and text-based access approaches to digital libraries through novel browsing, search, and visualization paradigms in addition to the traditional text search approach.

Project interactive approaches focused on general users with a straightforward retrieval paradigm that required the minimum retrieval effort from the user. As the majority of text searches were composed of 2-3 keywords, this was an important consideration.

The digital content used in the project was composed of television news, newspaper archives, museum photos, and personal digital photos. Material was supplied by the British Library, the Victoria & Albert Museum, BBC, University of Waikato, and Imperial College London.

• Developed a query-by-example mode using automated content-based analysis, which ultimately extracted salient visual and textual features from the multimedia objects and created corresponding indices.
• Explored new browsing paradigms that made use of the above similarity indices.
• Devised new search paradigms based on lateral browsing.
• Presented and summarized results adequately through document clustering, automated natural language summarization of text and speech, and storyboard generation of video material.
• Defined new interfaces that integrated and synchronized the different modes for resource discovery in digital libraries.

The project showcased novel search and browsing engines, allowing searches of multimedia collections in alternative ways, e.g., image similarity, as well as textual metadata, which may or may not have been present