HighFM: Towards a Foundation Model for Learning Representations from High-Frequency Earth Observation Data

Stella Girtsou, Konstantinos Alexis, Giorgos Giannopoulos, Charalambos Kontoes

Keywords: self-suprevised learning, foundation models, seviri, real-tim

Domains: Vision and Learning

External Link: https://arxiv.org/abs/2604.04306

Abstract:

The increasing frequency and severity of climaterelated disasters have intensified the need for realtime monitoring, early warning, and informed decision-making. Earth Observation (EO), powered by satellite data and Machine Learning (ML), offers powerful tools to meet these challenges. Foundation Models (FMs) have revolutionized EO ML by enabling general-purpose pretraining on large-scale remote sensing datasets. However most existing models rely on high-resolution satellite imagery with low revisit rates—limiting their suitability for fast-evolving phenomena and time-critical emergency response. In this work, we present HighFM, a first cut approach towards a FM for
high-temporal-resolution, multispectral EO data. Leveraging over 2 TB of SEVIRI imagery from the Meteosat Second Generation (MSG) platform, we adapt the SatMAE masked autoencoding framework to learn robust spatiotemporal representations. To support real-time monitoring, we enhance the original architecture with fine-grained temporal encodings to capture short-term variability. The pretrained models are then fine-tuned on cloud masking and active fire detection tasks. We benchmark our SEVIRI-pretrained Vision Transformers against traditional baselines and recent geospatial FMs, demonstrating consistent gains across both balanced accuracy and IoU metrics. Our results highlight the potential of temporally dense geostationary data for real-time EO, offering a scalable path toward foundation models for disaster detection and tracking. This paper has been accepted for publication at IJCAI 2026, Special Track On AI and Social Good.

Submission Number: 215