Panoramic image-based navigation for smart-phone in indoor environment

Abstract

In this paper, we propose a vision-based indoor navigation system for a smart-phone. The proposed system is designed to help a user traveling around an indoor environment to determine his current position and to give him the direction toward a chosen destination. For sensing user's position and orientation, the system utilizes panoramic images, which are pre-captured the environment and then processed to create a database. For matching images captured from user's smart-phone with the database, we use SURF[1], a robust detector and descriptor. Besides, to minimize responding time, the system employs client-server architecture in which a server module is mainly in charge of time consuming processes. Also, a tracking mechanism is applied to reduce matching time on the server. The experimental results show that the system can work well on a smart-phone in interactive time. © 2011 Springer-Verlag.