VidOnt–Spatiotemporal Video Annotation and Reasoning

VidOnt (VidOnt) is an OWL 2 ontology designed for spatiotemporal video annotation and reasoning for automated scene interpretation. It is one of the most expressive multimedia ontologies grounded in a hybrid description logic formalism and has a proper RBox. It features common video events for high-level descriptions with rich semantics, sets a new hierarchical structural for, and extends, standard spatial and temporal roles, and provides DL-safe rules.