Abstract

The diffuse availability of mobile devices, such as smartphones and tablets, has the potential to bring substantial benefits to the people with sensory impairments. The solution proposed in this paper is part of an ongoing effort to create an accurate obstacle and hazard detector for the visually impaired, which is embedded in a hand-held device. In particular, it presents a proof of concept for a multimodal interface to control the orientation of a smartphone's camera, while being held by a person, using a combination of vocal messages, 3D sounds and vibrations. The solution, which is to be evaluated experimentally by users, will enable further research in the area of active vision with human-in-the-loop, with potential application to mobile assistive devices for indoor navigation of visually impaired people.