Marvin Burges is a university assistant at the Computer Vision Lab at TU Wien, where he conducts research at the intersection of computer vision and machine learning. He received his Diploma in Computer Science (Dipl. Inf) from TU Dresden, specializing in computer vision. His diploma thesis focused on image descriptor learning for matching historical aerial images with present-day satellite images. Additionally, Marvin completed a PhD internship at Oak Ridge National Laboratory in Tennessee, USA. There, he collaborated with an interdisciplinary team of experts in computer vision, remote sensing, and human geography. Together, they developed a framework that combines active learning and foundation models to streamline dataset annotation for object detection in remote sensing imagery.
His current research focuses on combining machine learning with human-in-the-loop approaches to enhance model reliability and real-world applicability. As part of the interdisciplinary DoRIAH project, he collaborates with academic and industry partners to develop methods for semi-automatic detection of small objects in diverse remote sensing images. His contributions include implementing QGIS plugins to make complex models more accessible to domain experts and conducting user studies to ensure that his models are usable by non-expert users. Applications of this work range from detecting bomb craters in WWII-era aerial images to identifying vehicles in contemporary satellite images for traffic monitoring. Additionally, he works on domain adaptation and remote sensing applications, often involving human interaction to improve model performance. Looking ahead, he is interested in exploring physics-informed techniques as another form of grounding to further enhance model performance and applicability.
His current research focuses on domain adaptation, interactive or active learning-based object detection/segmentation in the general field of remote sensing.
Teaching
| Number | Type | Title | Semester |
| 183.130 | UE | 3D Vision | 2026S |
| 186.822 | VU | Introduction to Visual Computing | 2026S |
| 193.189 | VU | Machine Learning for Visual Computing | 2025W |
| 193.194 | EX | Computer Vision in Industry | 2025W |
| 183.130 | UE | 3D Vision | 2025S |
| 186.822 | VU | Introduction to Visual Computing | 2025S |
| 183.130 | UE | 3D Vision | 2024S |
| 186.822 | VU | Introduction to Visual Computing | 2024S |
| 183.605 | VU | Machine Learning for Visual Computing | 2023W |
| 193.125 | VU | Fundamentals of Computer Vision | 2023W |
| 186.822 | VU | Introduction to Visual Computing | 2023S |
| 193.125 | VU | Fundamentals of Computer Vision | 2023S |
| 183.605 | VU | Machine Learning for Visual Computing | 2022W |
| 186.822 | VU | Introduction to Visual Computing | 2022S |
Education
| 2021 | Diplom Informatiker. TU Dresden. Thesis title: Image Descriptor Learning for Matching Historical Aerial Images with Present-Day Satellite Images. Thesis advisor: Sebastian Zambanini, Robert Sablatnig |
| 2021 | ERASMUS exchange, Technische Universität Wien, Computer Vision Lab. |
| 2024 | PhD-Internship, Oak Ridge National Laboratory, Tennessee, USA. |
| Q1 2026 (expected) | Dr. Techn, Technische Universität Wien, Computer Vision Lab. Thesis title: Object Detection in Remote Sensing – Interactive and Active Learning-based Approaches for Historical and Very High-Resolution Imagery, Thesis Supervisor: Robert Sablatnig |
Publications
Burges, M., Zambanini, S., Sablatnig, R. “Exploring Learning-Based Approaches for Bomb Crater Detection in Historical Aerial Images“, In Proceedings of the OAGM Workshop 2022 (pp. 60–66), 2023.
Burges, M., Zambanini, S., Sablatnig, R. “Self-Supervised Learning for Historical Aerial Images“, Accepted for publication in Proceedings of the OAGM Workshop 2023.
Burges, M., Zambanini, S., Pirker, P. “CHAI: Craters in Historical Aerial Images“, In IEEE/CVF Winter Conference on Applications of Computer Vision, WACV 2024 (pp. 8241–8250), 2024.
Burges, M., Zambanini, S., Sablatnig, R. “Making archives searchable: Vision-language models for classification of historical aerial imagery.“, In Proceedings of the 3rd ACM SIGSPATIAL International Workshop on Searching and Mining Large Collections of Geospatial Data, GeoSearch 2024 (pp. 1–8), 2024.
Burges, M., Zambanini, S., Sablatnig, R. “Interactive Object Detection for Tiny Objects in Large Remotely Sensed Images.“, In IEEE/CVF Winter Conference on Applications of Computer Vision, WACV 2025 (pp. 4604–4713), 2025.
Burges, M., Dias, P.A., Woody, C., Walters, S., Lunga, D.D. “Interactive Rotated Object Detection for Novel Class Detection in Remotely Sensed Imagery.“, In IEEE/CVF Winter Conference on Applications of Computer Vision, WACV 2025 – Workshops (pp. 1129–1137), 2025.
Burges, M., Dias, P.A., Woody, C., Walters, S., Lunga, D.D. “Active Learning Meets Foundation Models: Fast Remote Sensing Data Annotation for Object Detection.“, In IEEE/CVF International Conference on Computer Vision, ICCV 2025 (pp. ??–??), 2025.
Talks
- “Semi-Automatic Object Detection in Historical Aerial Images with a Human in the Loop”, 2nd EuroSDR Workshop, Rome, December 2022