Speaker Bio

Begüm Demir is a Professor and Head of the Remote Sensing Image Analysis (RSiM) group at the Faculty of Electrical Engineering and Computer Science, Technische Universität Berlin, Germany. She received the PhD degree in 2010 in Electronic and Telecommunication Engineering from Kocaeli University, Turkey. She performs research in the field of processing and analysis of remote sensing images for Earth Observation with interdisciplinary approaches associated to remote sensing, machine learning and big data management. She was a recipient of a Starting Grant from the European Research Council with the project “BigEarth-Accurate and Scalable Processing of Big Data in Earth Observation” in 2017, and the “2018 Early Career Award” presented by the IEEE Geoscience and Remote Sensing Society. Dr Demir is a senior member of IEEE since 2016.

TITLE: Recent Trends in Foundation Models for Earth Observation and Pathways Forward
Abstract: This keynote focuses on the rapid evolution of Foundation Models for Earth Observation (EO) with particular attention to the critical challenge of model efficiency. It begins by critically examining the efficiency of existing EO foundation models, followed by a presentation of recent developments in computationally efficient models. Subsequently, the talk introduces REMSA (Remote-sensing Model Selection Agent) that is the first Large Language Model agent designed to automate EO foundation model selection from natural language queries. In conjunction with REMSA, the EO Foundation Model Database (which is the first structured, schema-guided resource covering more than 150 models trained across diverse EO modalities) is presented. The talk concludes by discussing potential pathways for future research in the context of EO foundation models and agents.