A Stereo-Based System with Inertial Navigation for Outdoor 3D Scanning

In this paper, we introduce a 3D scanning system consisting of a stereo camera combined with an inertial navigation system. We employ this system to explore the suitability of the standard 3D scanning pipeline when working with stereo range data of objects and architecture in an outdoor setting. Our range scanning system is very mobile without any mechanical actuators. We rely solely on the inertial navigation system for coarse registration with out any cumbersome manual procedure or scene dependent visual feature tracking. Our detail registration is a variant of the Iterative Closest Point algorithm. We introduce a set of heuristics that aims to produce reliable and repeatable convergence despite the diversity of the objects, and types of motions that may be executed by the operator during the scanning process. We also show that in an outdoor setting, range data based on classical binocular stereo can be effective for 3D modeling. Our system produces results of only slightly lower quality than an indoor triangulation scanner at a much lower system cost.