Open access
Jul 2026
OSM-CLIP: Enhancing Remote Sensing Image–Text Representation Learning with OpenStreetMap Data
OSM-CLIP is introduced, a framework that exploits the freely available, continuously growing annotations of OpenStreetMap to provide regionally scalable, patch-level supervision for remote sensing image-text learning and demonstrates that freely available geographic annotations can serve as a powerful source of supervi...
Alessio Pierdominici, R. Ricci, Mohammed Alruqimi et al.
· Applied Sciences · 0 citations