UniphorM: A New Uniform Spherical Image Representation for Robotic Vision

Abstract

In this article, we present a new spherical image representation, called uniform spherical mapping of omnidirectional images (UniphorM), and show its strong potential in robotic vision. UniphorM provides an accurate and distortion-free representation of a 360-degree image, by relying on multiple subdivisions of an icosahedron and its associated Voronoi diagrams. The geometric mapping procedure is described in detail, and the tradeoff between pixel accuracy and computational complexity is investigated. To demonstrate the benefits of UniphorM in real-world problems, we applied it to direct visual attitude estimation and visual place recognition (VPR), by considering dual-fisheye images captured by a camera mounted on multiple robotic platforms. In the experiments, we measured the impact of the number of subdivision levels of the icosahedron on the attitude estimation error, time efficiency, and size of convergence domain of an existing visual gyroscope, using UniphorM and three competing mapping algorithms. A similar evaluation procedure was carried out for VPR. Finally, two new omnidirectional image datasets, one recorded with a hexacopter, called SVMIS+, the other based on the Mapillary platform, have been created and released for the entire research community.