Markus Käppeler

	PhD Student
	Postal address:	Albert-Ludwigs-Universität Freiburg Technische Fakultät Robot Learning Lab Georges-Köhler-Allee 080 D-79110 Freiburg i. Br., Germany
	Office:	080-00-008
	Email:	kaeppelm@cs.uni-freiburg.de

About me

01/2024 - current: Ph.D. Student in the Robot Learning Lab, University of Freiburg
04/2020 - 12/2023: M.Sc. in Computer Science, University of Freiburg
10/2016 - 02/2020: B.Sc. in Applied Computer Science, Konstanz University of Applied Sciences

Research Interests

Robot Perception
Map-Based Perception
Foundation Models for Autonomous Driving
Self-Supervised and Label-Efficient Learning

Current Research Projects

AI-based System Architectures for Autonomous Driving

Teaching

TA, FreiCAR - Practical Autonomous Driving, University of Freiburg, WS2025
TA, Deep Learning Lab, University of Freiburg, SS2025
TA, Advanced Deep Learning, University of Freiburg, SS2025
TA, Robot Learning Seminar, University of Freiburg, WS2024/25
TA, Advanced Deep Learning, University of Freiburg, SS2024

Publications

2025

Markus Käppeler, Özgün Çiçek, Daniele Cattaneo, Claudius Gläser, Yakov Miron, Abhinav Valada,
Bridging Perspectives: Foundation Model Guided BEV Maps for 3D Object Detection and Tracking
arXiv preprint arXiv:2510.10287, 2025.
Paper Video Website BibTeX

@article{kaeppeler2025dvd,
      author={Käppeler, Markus and Çiçek, Özgün and Cattaneo, Daniele and Gläser, Claudius and Miron, Yakov and Valada, Abhinav},
      title={Bridging Perspectives: Foundation Model Guided BEV Maps for 3D Object Detection and Tracking}, 
      journal={arXiv preprint arXiv:2510.10287},
      year={2025},
}

Niclas Vödisch*, Kürsat Petek*, Markus Käppeler*, Abhinav Valada Wolfram Burgard,
A Good Foundation is Worth Many Labels: Label-Efficient Panoptic Segmentation
IEEE Automation and Robotics Letters (RA-L), vol. 10, no. 1, pp. 216-223, 2025.
Paper IEEE Xplore Video Website BibTeX

@ARTICLE{voedisch2025pastel,
  author={Vödisch, Niclas and Petek, Kürsat and Käppeler, Markus and Valada, Abhinav and Burgard, Wolfram},
  journal={IEEE Robotics and Automation Letters}, 
  title={A Good Foundation is Worth Many Labels: Label-Efficient Panoptic Segmentation}, 
  year={2025},
  volume={10},
  number={1},
  pages={216-223},
}

2024

Markus Käppeler*, Kürsat Petek*, Niclas Vödisch*, Wolfram Burgard, Abhinav Valada
SPINO: Few-Shot Panoptic Segmentation With Foundation Models
IEEE International Conference on Robotics and Automation (ICRA), 2024.,
Paper IEEE Xplore Video Website BibTeX

@inproceedings{kaeppeler2024spino,
    title={Few-Shot Panoptic Segmentation With Foundation Models},
    author={Käppeler, Markus and Petek, Kürsat and Vödisch, Niclas and Burgard, Wolfram and Valada, Abhinav},
    booktitle={IEEE International Conference on Robotics and Automation (ICRA)},
    year={2024},
    pages={7718-7724}
}