KI 2005 Workshop 7 Mixed-reality as a challenge to image understanding and artificial intelligence September 11 th , 2005 Koblenz

Knowledge representation and annotation of multimedia doc uments typically have been pursued in two different directions. Previous approac hes have focused either on low level descriptors, such as dominant color , or on the content dimension and corresponding manual annotations, such as personor vehicle. In this paper, we present a knowledge infrastructure to bri dge the gap between the two directions. Ontologies are being extend ed and enriched to include low-level audiovisual features and descriptors. Additionally, a too l f r linking low-level MPEG-7 visual descriptions to ontologies and annotations has been develope d. In this way, we construct ontologies that include prototypical instances of domain concepts tog ether with a formal specification of the corresponding visual descriptors. Thus, we combine high-l evel domain concepts and low-level multimedia descriptions, enabling for new media content an alysis.