Abstract
In this paper we formalise an ontology for motion verbs based on classical logic, event calculus, supervaluation and standpoint semantics. We present a theoretical account and a logic-programming implementation of the ontology, which fits within a system aimed at detecting event occurrences in video scenes. Our purpose is to build a bridge between Computer Vision and Knowledge Representation and Reasoning, and to address the issue of concept vagueness in formal ontologies. © 2012 Springer-Verlag.
Cite
CITATION STYLE
D’Odorico, T., & Bennett, B. (2012). Detecting events in video data using a formal ontology of motion verbs. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 7463 LNAI, pp. 338–357). https://doi.org/10.1007/978-3-642-32732-2_22
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
