FZJ‑Led Workshop Day at the HDCRS Summer School 2026

The High‑Performance and Disruptive Computing in Remote Sensing (HDCRS) Summer School 2026 will feature a full workshop day led by Forschungszentrum Jülich (FZJ) on 11 June 2026, focusing on Geospatial Foundation Models and hands‑on experimentation with TerraMind for downstream Earth observation tasks.  

The summer school is organised by the High‑Performance and Disruptive Computing in Remote Sensing (HDCRS) Working Group, part of the IEEE Geoscience and Remote Sensing Society (GRSS) within the Earth Science Informatics (ESI) Technical Committee. The programme brings together students, early‑career researchers, and senior experts working across remote sensing, high‑performance and distributed computing, disruptive computing approaches, and specialised hardware.

Rocco Sedona (FZJ) will open the workshop with an introduction to Geospatial Foundation Models for EO, providing an overview of current architectures, embedding strategies, and application domains. He will be joined throughout the day by Kennedy Adriko (FZJ) and Benedikt Blumenstiel (IBM).

Participants will then move to practical sessions designed to build proficiency with EO embedding models. Hands‑on TerraMind sessions will demonstrate how pretrained embeddings support downstream EO workflows, with extended exercises and concrete examples illustrating typical use‑case patterns.

The day will conclude with a dedicated session on the Embed2Scale FZJ‑led use case “Crop Stress and Yield Early Detection.” This use case already leverages the TerraMind geospatial foundation model, and the workshop will highlight recent progress and invite technical feedback. Kennedy Adriko will outline ongoing work to incorporate additional crops and explore adaptive fusion of multimodal GeoFM embeddings to support scalable yield and phenology research. The session will also touch on two active research directions: cross‑domain adaptation learning and assessing the transferability of embeddings across regions.