You Are Here: Geolocation by Embedding Maps and Images

TL;DR

This paper introduces a novel geolocation method that embeds panoramic images and maps into a low-dimensional space, enabling accurate route-based localization with high accuracy and fast convergence.

Contribution

It generalizes previous semantic feature methods by learning an embedded space, significantly improving localization accuracy and speed for route-based geolocation tasks.

Findings

Over 90% accuracy for 200m routes

Faster convergence compared to previous methods

Effective with Google Street View and Open Street Map data

Abstract

We present a novel approach to geolocalising panoramic images on a 2-D cartographic map based on learning a low dimensional embedded space, which allows a comparison between an image captured at a location and local neighbourhoods of the map. The representation is not sufficiently discriminatory to allow localisation from a single image, but when concatenated along a route, localisation converges quickly, with over 90% accuracy being achieved for routes of around 200m in length when using Google Street View and Open Street Map data. The method generalises a previous fixed semantic feature based approach and achieves significantly higher localisation accuracy and faster convergence.