Feature Matching Across 1D Panoramas

This paper presents a new method for feature matching between pairs of one-dimensional panoramic images for use in navigation and localization by a mobile robot equipped with an omnidirectional camera. We extract locally scaleinvariant feature points from the scale space of such images, and collect color information and shape properties of the scale-space surface in a feature descriptor. We define a matching cost based on these descriptors, and present a novel dynamic programming method to establish globally optimal feature correspondences between images taken by a moving robot. Our method can handle arbitrary rotations and large numbers of missing features. It is also robust to significant changes in lighting conditions and viewing angle, and in the presence of some occlusion.