Marian Himstedt

Associate Professor

prof_pic.png

Mönkhofer Weg 239

23562 Lübeck

Germany

I am an associate professor for computer science at the Technical University of Applied Sciences Lübeck, Germany.

My research is at intersection of computer vision, robotics and machine learning. I mainly focus on geometric vision and deep learning for navigation tasks such as simultaneous localization and mapping (SLAM), structure from motion (SfM) and more generally image registration. I currently work on applications of SLAM and SfM for endoscopy as well as registration for image guided therapies.

Before joining the TH Lübeck, Germany I worked as professor at the DHSH, as post-doc at the Medical Deep Learning Lab as well as Fraunhofer MEVIS. I also worked as R&D Engineer with SICK.

I completed a PhD in robotics focusing on semantic models for localization and mapping under supervision of Prof. Erik Maehle at the Institute of Computer Engineering of University of Lübeck, Germany in 2018.

news

Jul 8, 2025 The code for BronchoGAN has just been released on Github. Try synthesizing real bronchoscopy images from your own CT renderings and phantom recordings!
May 28, 2025 Our paper “BronchoGAN: anatomically consistent and domain-agnostic image-to-image translation for video bronchoscopy” has accepted for publication in IJCARS. We will also present at CARS in Berlin in June!
Mar 1, 2024 I started a new position as professor for computer science at the Technical University of Applied Sciences in Lübeck.
Jan 20, 2024 Our paper “Airway label prediction in video bronchoscopy: capturing temporal dependencies utilizing anatomical knowledge” has been published in IJCARS.
Nov 1, 2022 I started a new position as professor for computer science at the DHSH University of Applied Sciences in Kiel.

selected publications