In this paper we propose a series of novel lines of research emerging from the encounter of computer vision and formal ontologies. The main underlying idea is that a genuine visual knowledge can only emerge by the integration of the algorithmic and the semantic layer, which are typical of the two disciplines. We also present some early attempts at applying the proposed hybrid approach to tackle some important and still unsolved issues.
Statistical pattern recognition meets formal ontologies: Towards a semantic visual understanding
Contributo in atti di convegno
Roadmapping the Future of Multimodal Interaction Research including Business Opportunities and Challenges (RFMIR), pp. 23–25, Istanbul, 12-16/11/2014