Towards a Semantic Video Analysis using Deep Learning and Ontology

Abstract

With the technological advances in the field of multimedia, associated with the generalization of their uses in many applications such as television archiving, motion tracking, video surveillance, etc. Semantic analysis and automatic understanding of large collections of video documents become a major problem. Consequently, the need for a system, which will allow to effectively manipulate video content is undeniable. This paper presents an approach that allows the systematic video analysis using deep learning and ontology generation. The proposed approach permits the extraction and the building of an ontology using results obtained by the deep learning techniques such as key frames, detected objects and actions (movements).