OpenStreetView-5M: The Many Roads to Global Visual Geolocation

TL;DR

OpenStreetView-5M is a large, open-access dataset of over 5 million geo-referenced street view images from around the world, enabling more robust evaluation of visual geolocation algorithms.

Contribution

The paper introduces OpenStreetView-5M, a comprehensive dataset with strict train/test separation to improve evaluation of geolocation models.

Findings

Benchmarking of state-of-the-art encoders and strategies.

Dataset covers 225 countries and territories.

Supports evaluation beyond memorization.

Abstract

Determining the location of an image anywhere on Earth is a complex visual task, which makes it particularly relevant for evaluating computer vision algorithms. Yet, the absence of standard, large-scale, open-access datasets with reliably localizable images has limited its potential. To address this issue, we introduce OpenStreetView-5M, a large-scale, open-access dataset comprising over 5.1 million geo-referenced street view images, covering 225 countries and territories. In contrast to existing benchmarks, we enforce a strict train/test separation, allowing us to evaluate the relevance of learned geographical features beyond mere memorization. To demonstrate the utility of our dataset, we conduct an extensive benchmark of various state-of-the-art image encoders, spatial representations, and training strategies. All associated codes and models can be found at https://github.com/gastruc/osv5m.