Home video management system with semantic information

Abstract

The increasing amount of home video content has resulted in the need of efficient systems for the management of home video content. In this paper, we discuss the design of a home video management system that relies on the use of several semantic web technologies. The proposed system first segments video into shots. In a next step, semantic information is automatically extracted from key frames that are representative for the shots found. The extracted information is subsequently represented using the resource description framework (RDF), satisfying the constraints of a web ontology language (OWL) schema. The use of RDF facilitates interoperability among different sources of metadata and makes the information understandable for machines. Further, the use of RDF in our system design also allows using the SPARQL protocol and RDF query language (SPARQL) to query our metadata and to retrieve the video content the user is interested in.