Danish Airs and Grounds: A Dataset for Aerial-to-Street-Level Place Recognition and Localization

TL;DR

This paper introduces the DAG dataset, a large and diverse collection of aerial and street-level images designed to challenge and benchmark place recognition and localization methods under extreme viewpoint differences.

Contribution

The paper presents the DAG dataset with extensive aerial and street-level images and metadata, and proposes a map-to-image re-localization pipeline for improved localization accuracy.

Findings

The DAG dataset is larger and more diverse than existing datasets.

The proposed pipeline effectively matches street-level images to aerial-based 3D models.

Benchmark results demonstrate the dataset's utility for evaluating localization methods.

Abstract

Place recognition and visual localization are particularly challenging in wide baseline configurations. In this paper, we contribute with the \emph{Danish Airs and Grounds} (DAG) dataset, a large collection of street-level and aerial images targeting such cases. Its main challenge lies in the extreme viewing-angle difference between query and reference images with consequent changes in illumination and perspective. The dataset is larger and more diverse than current publicly available data, including more than 50 km of road in urban, suburban and rural areas. All images are associated with accurate 6-DoF metadata that allows the benchmarking of visual localization methods. We also propose a map-to-image re-localization pipeline, that first estimates a dense 3D reconstruction from the aerial images and then matches query street-level images to street-level renderings of the 3D model. The dataset can be downloaded at: https://frederikwarburg.github.io/DAG