Semantic annotation of soccer videos by visual instance clustering and spatial/temporal reasoning in ontologies

In this paper we present a framework for semantic annotation of soccer videos that exploits an ontology model referred to as Dynamic Pictorially Enriched Ontology, where the ontology, defined using OWL, includes both schema and data. Visual instances are used as matching references for the visual descriptors of the entities to be annotated. Three mechanisms are included to support effective annotation: visual instance clustering-to cluster instances of similar patterns, prototype selection-to select one or more visual representatives of each cluster, dynamic cluster updating-to update clusters and prototypes whenever new knowledge is presented to the ontology. Experimental results show the capability of performing semantic annotation of entities that exhibit a variety of complex changes in visual appearance or of events that show complex motion patterns in the same shot. SWRL rules are used to perform rule-based reasoning over both concepts and concept instances, to improve the quality of the annotation.