Title: A novel ontology for 3D semantics: ontology-based 3D model indexing and content-based video retrieval applied to the medical domain
Authors: Leslie F. Sikos
Addresses: Centre for Knowledge and Interaction Technologies, School of Computer Science, Engineering and Mathematics, Flinders University, Adelaide, Australia
Abstract: Because of the growing popularity of 3D modelling, there is a great demand for efficient mechanisms to automatically process 3D contents. Owing to the lack of semantics, however, most 3D scenes cannot be interpreted by software agents. 3D ontologies can provide formal definitions for 3D objects; however, many of them are semi-structured only, cover a narrow knowledge domain, do not provide comprehensive coverage for geometric primitives, and do not exploit the full expressivity of the implementation language. This paper presents the most comprehensive formally grounded 3D ontology to date that maps the entire XSD-based vocabulary of the industry standard X3D (ISO/IEC 19775-19777) to OWL 2, complemented by fundamental concepts and roles of the 3D modelling industry not covered by X3D. This upper ontology can be used for the representation, annotation, and efficient indexing of 3D models, and their retrieval by 3D characteristics rather than by associated category labels.
Keywords: 3D model semantics; multimedia ontology; medical 3D printing; feature-based 3D model retrieval; X3D; MPEG-7; medically accurate 3D models; content-based medical video retrieval.
International Journal of Metadata, Semantics and Ontologies, 2017 Vol.12 No.1, pp.59 - 70
Received: 17 Mar 2017
Accepted: 18 Aug 2017
Published online: 25 Oct 2017 *