You are here: Home Staff Jannik Zürn

Jannik Zürn

Albert-Ludwigs-Universität Freiburg
Technische Fakultät
Autonome Intelligente Systeme
Georges-Köhler-Allee 080
D-79110 Freiburg i. Br., Germany
Office:   080-00-008

Phone:   +49 761 203-8011
Fax:   +49 761 203-8007

This website is only updated periodically. For more information, please visit my personal homepage.

About me

  • I left the University of Freiburg and started a new chapter working at Wayve.
  • 10/2022 - 02/2023: Visiting PhD Student at Oxford Robotics Institute, University of Oxford (collaboration with Prof. Ingmar Posner)
  • 12/2018 - 11/2023: Working at the Autonomous Intelligent Systems group of Prof. Dr. Wolfram Burgard as Ph.D. student.
  • 07/2017 - 10/2017: Internship, Robotics Software Engineering at Mayfield Robotics, Redwood City, CA, US
  • 08/2015 - 10/2018: Master of Science in Theoretical Mechanical Engineering with minor in Computational Mechanics and Robotics, Karlsruhe Institute of Technology (KIT)
  • 10/2011 - 08/2015: Bachelor of Science in Mechanical Engineering with minor in Continuum Mechanics, Karlsruhe Institute of Technology (KIT)

Research Interests

For my PhD, I am mainly interested in autonomous urban navigation with multi-modal and self-supervised learning. In my research, I aim at bringing Robotics and Artificial Intelligence, especially Deep Learning, closer together. My goal is to enable autonomous robots to better understand their surroundings with the sensors they have and to allow them to more accurately and robustly navigate through those surroundings; especially in presence of adversarial influences such as sensor noise, uncertainties, and occlusion. Self-supervised learning with multiple sensor modalities plays a particularly important role in this endeavour as it allows us to avoid expensive and time-consuming labeling of data which is necessary for fully supervised learning.
  • Robot Perception
  • Weakly-Supervised Robot Learning
  • Self-Supervised Robot Learning
You can find my personal homepage here.
You can find my CV here.

Current Research Projects


Supervised Students

  • Semantic Segmentation of Curb and Curb Cuts in Street Imagery, Y. Satyawan, 2019, Bachelor Thesis
  • Multimodal Object Tracking with Deep Learning, T. Krautschneider, 2019, Bachelor Thesis
  • Optical Flow based Window Detection, G. Stief, 2020, Bachelor Thesis
  • Sound Event Localization and Detection, S. Al-Rawi, 2021, Master Thesis
  • Self-Supervised Road Crossing Detection for Pedestrian Robots, S. Weber, 2022, Master Thesis


Conference Papers

  • Martin Büchner*, Jannik Zürn*, Ion-George Todoran, Abhinav Valada, Wolfram Burgard
    NEW: Learning and Aggregating Lane Graphs for Urban Automated Driving
    IEEE / CVF Computer Vision and Pattern Recognition Conference (CVPR), 2023.
    Download Website BibTeX
  • Jannik Zürn*, Sebastian Weber*, Wolfram Burgard
    TrackletMapper: Ground Surface Segmentation and Mapping from Traffic Participant Trajectories
    Conference on Robot Learning (CoRL), 2022.
    Download Website BibTeX
  • Jannik Zürn, Wolfram Burgard
    Self-Supervised Moving Vehicle Detection from Audio-Visual Cues
    IEEE Robotics and Automation Letters | IEEE International Conference on Intelligent Robots and Systems (IROS), 2022.
    Download Website BibTeX
  • Johan Vertens*, Jannik Zürn*, Wolfram Burgard
    HeatNet: Bridging the Day-Night Domain Gap in Semantic Segmentation with Thermal Images
    IEEE International Conference on Intelligent Robots and Systems (IROS), 2020.
    Download Website BibTeX

Journal Articles

  • Jannik Zürn, Ingmar Posner, Wolfram Burgard
    NEW: AutoGraph: Predicting Lane Graphs from Traffic Observations
    Arxiv Preprint 2306.15410
    Download Website BibTeX
  • Jannik Zürn*, Johan Vertens*, Wolfram Burgard
    Lane Graph Estimation for Scene Understanding in Urban Driving
    IEEE Robotics and Automation Letters, 2021.
    Download Website BibTeX
  • Jannik Zürn, Wolfram Burgard Abhinav Valada
    Self-Supervised Visual Terrain Classification from Unsupervised Acoustic Feature Learning
    IEEE Transactions on Robotics, 2021.
    Download Website Video BibTeX


Benutzerspezifische Werkzeuge