Earth Embeddings: Learning Mental Maps in Neural Nets
Year: 2025
Type: article
Abstract: What makes AI models robust when reasoning about geography, climate, or human activity? While explicit geospatial data—like temperature, precipitation, or land cover—can inform models, it’s often incomplete, sparse, or too coarse to generalize well. Instead, we explore how to create implicit, robust environmental embeddings that capture the complex and diverse character of places directly from imagery. This talk introduces a novel approach to learning these embeddings by training AI to play Satellite GeoGuessr—contrastively learning what makes each location unique through satellite images. We present two complementary methods: First, we introduce location encoders—neural networks that learn to map coordinates to rich geospatial representations, storing knowledge in their weights. Second, we describe SatCLIP Earth Embeddings, which trains image and location encoders jointly to produce embeddings that reflect both the visual and contextual identity of a place.
DOI: 10.22032/dbt.67647
Authors: Friedrich-Schiller-Universität Jena. Multimediazentrum, ELLIS Unit Jena, Rußwurm, Marc
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7090184587
FWCI:
Cites:
Cited by:
Related to: