Learning to Map Nearly Anything

TL;DR

This paper introduces a novel cross-modal distillation method enabling the estimation of fine-grained land properties from overhead imagery without manual annotations, improving mapping and localization tasks.

Contribution

It presents an extensible approach that leverages cross-modal distillation to predict detailed land properties from overhead images without manual labeling.

Findings

Effective prediction of fine-grained land properties from overhead imagery.

Models improve mapping and localization applications.

No manual annotation required for training.

Abstract

Looking at the world from above, it is possible to estimate many properties of a given location, including the type of land cover and the expected land use. Historically, such tasks have relied on relatively coarse-grained categories due to the difficulty of obtaining fine-grained annotations. In this work, we propose an easily extensible approach that makes it possible to estimate fine-grained properties from overhead imagery. In particular, we propose a cross-modal distillation strategy to learn to predict the distribution of fine-grained properties from overhead imagery, without requiring any manual annotation of overhead imagery. We show that our learned models can be used directly for applications in mapping and image localization.