Distributed cross-modal search within the MPEG query format

One of the latest developments of the MPEG committee is the Query Format for search and retrieval of multimedia content. This language constitutes the interface between a client and a search engine for searching multimedia data. Another possible scenario is the use of a service provider, which accepts and understands a query from a client and forwards parts of the query to one or more specific databases. Furthermore the service provider is able to retrieve the reply from these databases and post processes this result in order to send it to this client. During the last years, the cross-modal search of video and audio signals became more and more important, since using both, the video and the audio signal together turned out to be much more robust for identification of video streams, than the image part of the video alone.This paper describes a cross-modal search based on the MPEG Query Format, using a service provider for splitting and aggregating the query and two different types of databases.