The Remote Sensing Technology Institute is a DLR institute with sites in Oberpfaffenhofen near Munich, Berlin-Adlershof and Neustrelitz in Mecklenburg-Western Pomerania. Together with the German Remote Sensing Data Center, the institute forms the Earth Observation Center EOC, the competence center for earth observation in Germany.

What to expect:

The Department of Photogrammetry and Image Analysis at the Remote Sensing Technology Institute is concerned with the development of operational image processors for the extraction of geometric (photogrammetry) and semantic (image analysis) information from digital remote sensing images from satellites, aircraft and UAVs. The methodological and application-oriented scientific developments for the automatic extraction of geoinformation from optical or multimodal data sets include the following research focuses of the department: stereo processing and object recognition and real-time methods for image data from aerial image sensors for traffic monitoring.

As part of your studies, we offer you the opportunity to write a master thesis on the topic: 
Deep Learning-based Building Attribute Estimation from Street-Level Imagery
Semantic building information—such as function, number of floors, roof shape, and construction epoch—is essential for various applications but often missing from building databases. This thesis aims to estimate multiple building attributes using street-level imagery. The objective is to develop a robust building attribute dataset that can serve as a benchmark and to train baseline models for attribute estimation. The resulting models should ideally generalize beyond the original dataset region to other areas across Germany or Europe, providing added value to urban planning and development stakeholders.

Key aspects of the project include:

•    Processing a large-scale image dataset 
•    Training deep learning models to predict selected building attributes via a multi-task learning approach
•    Inferring missing building information where it is not available in existing database

Your Qualifications:
•    Ongoing Master’s studies in Computer Science, Geoinformatics, Data Science, AI, Remote Sensing, or a related field
•    Strong interest in AI, computer vision, and urban analysis
•    Experience with deep learning frameworks (e.g., PyTorch or TensorFlow) is a plus
•    Motivation and ability to work independently
•    Good communication skills and proficiency in English (spoken and written)

What we offer you:
•    A relevant and current research topic with societal impact
•    Supervision by experienced researchers
•    Flexible working hours and the option to work in a hybrid setup
•    Integration into a supportive, interdisciplinary research team
•    Compensation according to public service wage agreements (TVöD Bund, E5 for Master’s students)

DLR stands for diversity, appreciation and equality for all people. We promote independent work and the individual development of our employees in their personal and professional environment. Our numerous training and development opportunities are available to you for this purpose. Equal opportunities are particularly important to us, which is why we want to increase the proportion of women in science and management in particular. We give preference to applications from people with severe disabilities if they are professionally qualified.

We look forward to getting to know you!