Cooperative depth extraction

Abstract

In this paper we describe an approach to efficient and robust depth extraction using a cooperative sensor system for an autonomous mobile robot navigation system operating in static indoor environments. The cooperative sensor system comprises two parts: binocular region based stereo, i.e., a passive depth extraction technique, and depth from focus, i.e., an active depth extraction technique. The region based stereo technique provides an overview of the scene aided by a 3D a priori world model. Since feature based stereo is vulnerable to occlusions, the depth from focus technique is selectively employed at locations where potential occlusions are detected in order to extract the correct depth. Experimental results demonstrates the advantages of the cooperative system. 1 Introduction Efficient and robust depth extraction is an active research area in computer vision. This is due to two things: depth is the missing dimension in standard 2D imagery and the perception of 3D is in general c...