Blog
Notes from Locamage
What we're learning about reading the ground from satellites, planes and drones.
What changed
Finch follows farm fields in France through a season of Sentinel-2 imagery, and maps crops from radar alone.
What Finch sees
Segmentation maps from four public benchmarks, from 20 cm aerial photos to 30 m satellite imagery, all read by one frozen model.
Finch 1: one model for every sensor
Finch 1 is our first geospatial foundation model, trained on imagery from more than 50 sensors across satellites, aircraft and drones.