IN FOCUS: 2018 Early Career Awardee Begüm Demir

By Joanne Van Voorhis

IEEE GRSS supports a wide range of awards , many of which have a nomination deadline of December 15. We continue our series highlighting IEEE GRSS award winners by profiling Dr. Begüm Demir, who was named the 2018 Gold Early Career Awardee for her research contributions in machine learning for information retrieval in remote sensing. The award is a distinction that recognizes early-career researchers who have already demonstrated exceptional ability and show promise for significant future contributions to the society’s fields of interest. Demir is a Full Professor at Technische Universität Berlin (TU Berlin) in Germany, where she serves as the founder and head of the Remote Sensing Image Analysis (RSiM) research group within the Faculty of Electrical Engineering and Computer Science. In addition, she leads the Big Data Analytics for Earth Observation research group at the Berlin Institute for the Foundations of Learning and Data (BIFOLD).

Established in the early 2010s, IEEE GRSS Early Career Award recognizes excellence in research, innovation, and leadership, with recipients selected by a GRSS committee based on the originality and impact of their work, community engagement, and potential for future contributions. Presented annually at the IEEE International Geoscience and Remote Sensing Symposium (IGARSS), the award is one of the society’s highest distinctions for emerging researchers and serves as both recognition of early achievement and encouragement for continued advancement in Earth observation science and technology.

Recognition of Success and Inspiration for the Future

Begüm Demir’s selection for the 2018 Early Career Award reflected a sustained record of influential research at the intersection of machine learning, signal/image processing and large-scale Earth observation data analysis. She was recognized for her contributions in machine learning for information retrieval and scalable processing of big Earth observation (EO) data, as well as for her leadership in related research projects. At the time of the award she had built an international profile for novel methods in content-based remote sensing image retrieval, scalable classification and information discovery from very large satellite image archives. In the ensuing years, she has expanded on her prior work, and continues to make significant contributions to the field of remote sensing. Demir’s record speaks clearly: she has helped move the community toward more scalable, machine-learning-driven ways of turning satellite imagery into information.

“This award was very significant for me because it represents the culmination of years of hard work and dedication in the field of machine learning for remote sensing,” explains Demir. “It also served as an inspiration to strengthen my pursuit of research excellence, while also committing to mentor and foster the growth of the next generation of researchers in this important research field,” she adds.

Looking to the Future of Remote Sensing Technology

Demir is currently tackling several frontier research challenges at the intersection of machine learning, remote-sensing and big-data management. Her work centres on developing foundation models for Earth observation (EO) that handle vast, multi-modal satellite-image archives. In October, she presented an IEEE GRSS webinar and shared the RSiM group’s recent advances in foundation models for Earth observation (EO). She highlighted their EO Foundation Model Database – the first structured, schema-guided resource cataloging over 150 EO foundation models across diverse data modalities and learning paradigms. She also introduced REMSA (Remote-sensing Model Selection Agent), the first LLM agent for automated foundation model selection via natural language queries.

She is also advancing data-centric AI for EO by developing crucial benchmark datasets (e.g., BigEarthNet, HySpecNet-11K, TreeSatAI, upcoming BigEarthNet.txt). Further, Demir and colleagues are exploring multi-modal federated learning systems (for the analysis of remote sensing images across decentralized and unshared EO archives) and label-noise robust AI models (to reduce the negative impact of noisy land-use and land-cover annotations) for the accurate analysis of EO data. Explore more about Demir’s efforts here.

“I’ve recently focused on exploring vision-language models (VLMs) for EO,” says Demir. “I think this is particularly interesting since it offers a paradigm shift: unified models that understand both image content and natural language, enabling flexible EO data analysis through text-based interaction,” she adds.

Inspiring Others through Collaboration

Demir has been very active within the IEEE GRSS community. Her roles within IEEE GRSS span technical program leadership, editorial contributions, outreach/education via chapters and webinars, and community service – a blend of activities that reflect both her research stature and her engagement with the society’s mission. For example, she served as an Associate Editor for the GRSS journal IEEE Geoscience and Remote Sensing Letters (GRSL) until 2025, was appointed to the Editorial Board of the society’s magazine (the GRSS magazine) as an Associate Editor in 2024, and frequently presents webinars on a range of topics. At IGARSS 2025, she organized a session on Data-Centric AI in Geoscience Applications together with her colleagues at NASA.

“My academic journey has been strongly linked with the IEEE GRSS,” explains Demir. “I volunteer my time and expertise to support IEEE GRSS through which I have grown as a researcher, academic supervisor and professional. I encourage others to volunteer as well (e.g., proposing a community contributed session for IGARSS, acting as a Reviewer at GRSS journals, etc). The benefits are broad and whatever you can do can only make the global GRSS stronger,” she encourages.

Nominations for 2026 Awards Accepted until December 15

Do you know of someone who deserves recognition? We encourage you to consider nominating someone for a IEEE GRSS award. Explore the various awards and help us identify those who deserve recognition! Nominations (nt.grss-ieee.org/awards/) for most 2026 awards are accepted through December 15. Your participation in the process is appreciated as it helps us select from a large group of deserving nominees from all over the world.