Detecting privacy in attention aware system
J. Maisonnasse, Nicolas Gourier, Oliver Brdiczka, Patrick Reignier, James L. Crowley · 2006
Privacy and security are important issues for acceptation of ambient technology. For this reason, we propose a detector of relations between users and objects based on attention aware system. This detector is inspired by social sciences and completes the representation-driven model mainly used in ontology-based approaches. We discuss the role of context in these approaches and managing privacy. An attention model is derived from the gravitation model and cognitive psychology approaches. This model exploits contextual elements such as position, speed and saliency of objects in a scene to estimate shared attention. An application has been developed to demonstrate the efficiency of this system when managing windows applications, while guaranteeing the security and privacy of users. (9 pages)