GeoSynth: Contextually-Aware High-Resolution Satellite Image Synthesis
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866911834177536000 |
|---|---|
| author | Sastry, Srikumar Khanal, Subash Dhakal, Aayush Jacobs, Nathan |
| author_facet | Sastry, Srikumar Khanal, Subash Dhakal, Aayush Jacobs, Nathan |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2404_06637 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | GeoSynth: Contextually-Aware High-Resolution Satellite Image Synthesis Sastry, Srikumar Khanal, Subash Dhakal, Aayush Jacobs, Nathan Computer Vision and Pattern Recognition 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. |
| title | GeoSynth: Contextually-Aware High-Resolution Satellite Image Synthesis |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2404.06637 |