Generative domain adaptation and foundation models for robust Earth observation

Welcome to this week’s Learning Machines seminar.

This seminar is a collaboration between RISE and Climate AI Nordics – climateainordics.com.

Title: Generative domain adaptation and foundation models for robust Earth observation

Speaker: Georges Le Bellier, Conservatoire National des Arts et Métiers

Abstract: Deep learning for remote sensing plays a crucial role in turning satellite and aerial imagery into dependable, real-world insights. However, Earth observation models must handle diverse environments, sensors, and conditions—such as clouds, seasonal shifts, and geographic differences—while still producing accurate results. In this talk, we explore two paths that lead to more robust and adaptable algorithms: generative domain adaptation and geospatial foundation models.

First, I will introduce FlowEO, a generative approach of Unsupervised Domain Adaptation (UDA) for Earth observation, and show its high performance in UDA scenarios for several downstream tasks, including dense prediction and classification. This flow-matching-based translation method improves pretrained predictive models' accuracies in challenging scenarios such as post-disaster response and high cloud coverage cases with SAR-to-optical translation. FlowEO’s generative domain adaptation method is independent of the downstream task and does not require retraining the predictive model.

Then, I will present “PANGAEA: A Global and Inclusive Benchmark for Geospatial Foundation Models“, a standardized evaluation protocol that covers a diverse set of datasets, dense prediction tasks, resolutions, sensor modalities, and temporalities. This benchmark includes comparison between geospatial foundation models but also with supervised baselines, namely U-Net and ViT, and highlights the strengths and weaknesses of GFMs. In addition, PANGAEA evaluates models’ accuracy in cases where labels are limited and questions the impact of multi-temporal data for GFMs.

About the speaker: Georges Le Bellier is a third-year PhD candidate at the Conservatoire National des Arts et Métiers, where he researches generative models for Earth Observation. Previously, he explored generative approaches for controllable musical performance at Sony and IRCAM. He now focuses on developing efficient conditional generative models tailored to remote sensing. His current research combines flow-based methods with domain adaptation, aiming to enhance the robustness of EO models to shifting data distributions.

Location: This is an online seminar. Connect using Zoom.

Date: 2025-09-11 15:00

Upcoming seminars:

• 2025-09-25, 15:00: Sigrid Passano Hellan, NORCE Norwegian Research Centre
• 2025-10-09, 15:00: Sebastian Hafner, RISE
• 2025-10-23, : Nora Gourmelon, Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU), **digital and physical: **
• 2025-11-06, 15:00: Felix Koehler, TU Munich
• All times are in CET.

More information and coming seminars: https://ri.se/lm-sem

– The Learning Machines Team