Designing a modular system for direct search of video & audio content: The DIVAS project

The principal objective of the DIVAS IST project is to integrate essential knowledge from universities, research institutes, industrial partners and end-users, in order to design a multimedia search engine based on direct video and audio search algorithms applied on encoded representation formats of the content inside multimedia databases. The driving force is to disassociate content search from the availability of laboriously annotated metadata databases. Thus, developed algorithms provide an alternative and complementary path for metadata based audio/video content search. In this paper, the recent advances in the design and implementation of the DIVAS reference system are presented, giving rise to the distinct components that comprise the core framework. In addition, the driving force for developing a module-based flexible architecture that facilitates the ongoing expansion of the system based on researcherspsila contributions is also reported.