Wide-Area Image Geolocalization with Aerial Reference Imagery

TL;DR

This paper introduces a deep learning approach for cross-view image geolocalization using aerial and ground-level images, achieving significant improvements on benchmark datasets through novel feature representations and multi-scale fusion.

Contribution

It presents a new deep neural network architecture with cross-view training and multi-scale feature fusion for improved geolocalization accuracy.

Findings

Outperforms previous methods on benchmark datasets

Features are discriminative at local and continental scales

Introduces a large database of paired aerial and ground images

Abstract

We propose to use deep convolutional neural networks to address the problem of cross-view image geolocalization, in which the geolocation of a ground-level query image is estimated by matching to georeferenced aerial images. We use state-of-the-art feature representations for ground-level images and introduce a cross-view training approach for learning a joint semantic feature representation for aerial images. We also propose a network architecture that fuses features extracted from aerial images at multiple spatial scales. To support training these networks, we introduce a massive database that contains pairs of aerial and ground-level images from across the United States. Our methods significantly out-perform the state of the art on two benchmark datasets. We also show, qualitatively, that the proposed feature representations are discriminative at both local and continental spatial scales.