An AI Foundation Model to better understand our planet

As AI is revolutionising the way we analyse data, a team from École Polytechnique Fédérale de Lausanne (EPFL), Switzerland is studying how to create the first language-powered, sensor-agnostic AI Foundation Model that would be able to analyse independently remote sensing images from different Earth Observation (EO) satellites.

The model has been trained to extract information from multiple satellites (for example Sentinel-1 and Sentinel-2) over the same area, showing it can interpret new input efficiently and perform tasks beyond its training data sets.

Find out more here.