Advanced Multimedia Indexing and Searching
Project code: AMIS
Research group: Database Architectures and Information Access (INS1)
Project coordinator: Martin Kersten
The AMIS project is a national research initiative to broaden and deepen the understanding of methods for indexing and searching in multimedia databases. The methods should be both effective and efficient. In this project, several disciplines of computer science collaborate to achieve this one research goal of advancing insight into critical bottlenecks of multimedia technology.
To that end, consider the ground level of multimedia databases where the topic is data structures, given the storage technology as well as the query and index optimization. Key elements here are the proper and anticipating organization of the data in the database in order to answer any query quickly. In this respect, multimedia databases are distinctly different from the common categoric databases. The performance at this level is measured by response time.
The second level deals with formal descriptions of the semantic content as the basis for explicit querying, as well as the expression of the domain in invariant descriptions, enabling querying by example. At this level, knowledge of the search domain is integrated in methods. This is the level where the performance is measured in terms of precision and recall.
At the third level, the system is concerned with bringing in the data on time. The delivery of data 'just in time' plays a key role. Here guaranteed constant play out rates and low data latencies of data rendered on user displays play a key role. This is the level where performance is measured in terms of percieved 'quality of service' at display.
Succesfull tackling of the research issues requeries a better understanding of the interplay between multimedia data organization and the user in querying over formalized and unformalized multimedia domains.
Leading groups in The Netherlands in these fields have been brought together to undertake a concerted research programme to address these issues within the context of a focused application domain: a database for pictures, and video.
Members
Paul de Bra, Martin Kersten, Menzo Windhouwer
Cooperation
This project is financed by Technologiestichting STW.
