As deep learning revolutionizes Earth Observation analysis, Foundation Models emerge as a promising alternative to supervised learning, overcoming data labeling limitations by leveraging large-scale, self-supervised learning. The FM4CS project is developing a multi-modal Foundation Model for the European Space Agency, using multi-spectral and radar data from Sentinel satellites to enhance Earth Observation capabilities.
Arnt Børre Salberg: Foundation Models for Climate and Society
Arnt-Børre Salberg is a Chief Research Scientist at the Norsk Regnesentral, affiliated with the Department of BAMJO (Image Analysis, Machine Learning and Earth Observation).
His research primarily focuses on the developing deep learning, computer vision, and remote sensing methods to image data. This includes work on forest mapping and monitoring, marine science applications, coastal habitat mapping, foundation models, and analysis of data from various sensors like satellites, seismic, and marine echosounders.
Published Mar. 11, 2025 3:40 PM
- Last modified Apr. 4, 2025 2:46 PM