GeoSynth: Contextually-Aware High-Resolution Satellite Image Synthesis
This work addresses the need for customizable satellite image synthesis for applications like urban planning or simulation, but it appears incremental as it builds on existing generative models with added controls.
The paper tackles the problem of synthesizing satellite images with control over global style and layout, using textual prompts or geographic location, and demonstrates that the model generates diverse, high-quality images with excellent zero-shot generalization.
We present GeoSynth, a model for synthesizing satellite images with global style and image-driven layout control. The global style control is via textual prompts or geographic location. These enable the specification of scene semantics or regional appearance respectively, and can be used together. We train our model on a large dataset of paired satellite imagery, with automatically generated captions, and OpenStreetMap data. We evaluate various combinations of control inputs, including different types of layout controls. Results demonstrate that our model can generate diverse, high-quality images and exhibits excellent zero-shot generalization. The code and model checkpoints are available at https://github.com/mvrl/GeoSynth.