抄録
Human access to WWW information has to become media-independent to permit full access to content. For instance we need to have an image as a possible query in order to retrieve text, or a speech utterance as a query in order to retrieve motion image. We have developing a system called the CrossMediator which realizes a complex of retrieval software modules which works among multi-media databases composing of still image, motion image, speech and text.This paper shows some examples of the multi-media reterieval system.
