Tile Tracker: A Practical and Inexpensive Positioning System for Mobile AR Applications

Abstract

In this paper, we propose a practical and low-cost positioning system that quickly determines the camera pose in an environment with pre-existing ground tiles, with the support of fiducial images. Such environment exists in many places, both indoors and outdoors, such as museums and parks. Our system is designed to be used on mobile devices (e.g. smartphone, tablets, etc.) with inexpensive cameras to fast calibrate cameras and to track the movement of a user. Contrary to other existing mobile tracking systems, our algorithm has minimal drift over time. To accomplish this, our approach, fully taking advantage of the existing rectangular tiles, labels them with markers that supports quick orientation determination and unique identification. Our algorithm recovers the camera pose based on the images taken of these marked tiles in real time, and thus provides basis for a wide range of applications such as navigational assistance and augmented or virtual reality.