A Service-oriented Framework for Context-aware Semantic Media Applications

As we move closer to the vision of ubiquitous, invisible computing, we feel the need for a framework that supports the integration of visual and auditory inputs with context sourced from various ambient 'sense organs' such as temperature and pressure sensors etc, as well as other factual information about persons, locations, activity schedules and events that are of significance to human beings. In this paper we describe our framework for integrating visual inputs at the human-semantics level, so that these inputs can be used for reasoning and inference. We borrow techniques from media and knowledge representation as well as work in pervasive context-aware computing to build a unifying service-oriented framework that can support applications with differing needs.