Applying Media Semantics Mapping in a Non-linear, Interactive Movie Production Environment

In this work we propose how to deal with the Semantic Gap in closed domains. That is, we propose to bridge the Semantic Gap by means of mapping well- known low-level feature patterns found in MPEG-7 descriptions to formal concepts. The key contributions of the proposed approach are (i) the utilisation of ontologies, and rules to enhance the retrieval capabilities (eectiveness), and (ii) the realisation of the feature matching process being carried out on the structural level through indexed MPEG-7 descriptions (eciency). We discuss advantages and shortcomings of our approach, and illustrate its application in the realm of non-linear, interactive movie productions.