Generate Your Own Scotland: Satellite Image Generation Conditioned on Maps
This work addresses the underexplored application of diffusion models in Earth Observation, providing a tool for remote sensing researchers to generate synthetic satellite imagery, though it is incremental as it adapts existing methods to a new domain.
The authors tackled the problem of generating realistic satellite images from cartographic data by conditioning a pretrained diffusion model on OpenStreetMap images, resulting in a model that achieves both high image quality and map fidelity, as demonstrated through qualitative evaluation on two large datasets covering Mainland Scotland and the Central Belt.
Despite recent advancements in image generation, diffusion models still remain largely underexplored in Earth Observation. In this paper we show that state-of-the-art pretrained diffusion models can be conditioned on cartographic data to generate realistic satellite images. We provide two large datasets of paired OpenStreetMap images and satellite views over the region of Mainland Scotland and the Central Belt. We train a ControlNet model and qualitatively evaluate the results, demonstrating that both image quality and map fidelity are possible. Finally, we provide some insights on the opportunities and challenges of applying these models for remote sensing. Our model weights and code for creating the dataset are publicly available at https://github.com/miquel-espinosa/map-sat.