Ontology-Based Unification of MPEG-7 Semantic Descriptions

摘要

MPEG-7 provides a rich set of tools for describing multimedia content. Among them, the semantic descriptor scheme (DS) is used to describe the semantics of the content in terms of events, objects, concepts, places, time and abstraction. However, to describe similar facts presented in different multimedia contents, different semantic descriptions may result, even after following the MPEG-7 semantic DS. These semantic descriptions may be unified in order to enhance the overall knowledge about the associated multimedia content. This paper proposed an ontology that semantically represents the structure of MPEG-7 semantic DS and acts primarily as a resource of such unification. The knowledge representation provided by this ontology can be used to develop tools that perform automatic multimedia reasoning from different existing semantic descriptions, which are narrated with other domain specific ontologies.