Welcome to RSiM
Remote Sensing Image Analysis Group


The Remote Sensing Image Analysis (RSiM) group at the Faculty of EECS, TU Berlin and the Big Data Analytics for Earth Observation (BigEarth) group at BIFOLD are led by Prof. Begüm Demir. Our groups perform research in the field of processing and analysis of remote sensing images for Earth observation with interdisciplinary approaches associated to remote sensing, machine learning, signal & image processing and big data management.

Latest News

Post thumb
  • 2023-08-01

RSiM & BigEarth at IGARSS 2023

We presented 7 papers at IGARSS 2023 in Pasadena, California, USA. To read our IGARSS papers, please visit our publications page.

Post thumb
  • 2023-06-25


We have recently released HySpecNet-11k that is a large-scale hyperspectral benchmark dataset made up of 11,483 image patches acquired by EnMAP satellite. For details, please visit: hyspecnet.rsim.berlin.

Post thumb
  • 2023-06-08

Summer School in Cremona

We organized a summer school on "Machine Learning and Data Fusion for Earth Observation" in Cremona, Italy in collaboration with the University of Pavia, Grenoble Institute of Technology, University of Iceland.

Post thumb
  • 2023-05-09

Gencer Sumbul defended his PhD thesis

Gencer Sumbul successfully defended his PhD thesis titled “Deep Image Representation Learning for Knowledge Discovery from Earth Observation Data Archives” on May 9th and received the distinction 'summa cum laude'. Congratulations Gencer!

Post thumb
  • 2023-05-01


We have recently released ConfigILM that is a pytorch based library, enabling fast development of visual question answering systems. This open-source library provides a convenient implementation for seamlessly combining models from two of the most popular pytorch libraries, timm and huggingface.

Post thumb
  • 2023-01-11

Leonard Hackel won Rolf-Niedermeier-Award

Leonard Hackel was awarded the Rolf-Niedermeier-Award for the best master thesis in Computer Engineering of TU Berlin. In his thesis, he developed lightweight transformer-based visual question answering (VQA) models for Earth observation. For details, click here.

Supported by

logo logo logo logo logo