SURFtogether: Towards Context Proximity Detection Using Visual Features

Abstract

With the now near ubiquity of smart mobile devices and the advent of new wearable computing devices like Google Glass, context-aware computing applications are becoming more and more feasible. We propose a new concept coined Context-Proximity Awareness aimed at identifying closely related entities based on contextual similarity. As a first step towards that goal, we introduce the SURFtogether approach, trying to detect contextual proximity by analyzing the field of vision of two or more entities. We evaluate the general feasibility of our approach based on real-world data and show that in our initial tests, correct detection of contextual proximity is achieved nearly 90p of the time.