Semi-automatic extraction of semantics from football video sequences

Abstract

In this paper we use a knowledge based inference model in combination with knowledge stored in a domain specific ontologies to organize and combine semantically-related entities, within multimedia documents in order to extract semantics from video sequences. The simulation results demonstrate that is possible to extract high-level information from video if low-level features are extracted with high precision. Within the scope of this work we are not particularly interested in the method by which the low level features are extracted from the video.