Jan-Aike Termöhlen
PhD Student focused on machine learning for domain adaptation and generalization in computer vision
About
PhD Student in electrical engineering at the Institute for Communications Technology (TU Braunschweig), specalizing in machine learning for computer vision. My main research interests lie in deep learning methods for unsupervised domain adaptation, continual domain adaptation, and domain generalization, with the goal to bridge the domain gap between synthetic and real data for neural network training and inference.
Work Experience
Research Assistant
Institute for Communications Technology, TU Braunschweig
Student Research Assistant
Institute for Communications Technology, TU Braunschweig
Student AssistantPart Time
Institute of Electrical Metrology and Fundamentals of Electrical Engineering, TU Braunschweig
Intern
Siemens AG
Team LeaderPart Time
StudING - Das studentische Ingenieurbüro UG
Education
PhD Student in Machine Learning and Computer Vision
M.Sc. in Electrical Engineering
B.Sc. in Industrial Engineering specialised in Electrical Engineering
Programming / Software Skills
Languages
Publications
Generalization by Adaptation: Diffusion-Based Domain Extension for Domain-Generalized Semantic Segmentation
Joshua Niemeijer*, Manuel Schwonberg*, Jan-Aike Termöhlen*, Nico M. Schmidt, and Tim Fingscheidt (* equal contribution)
In Proc. of WACV, pages 2830–2840, Waikoloa, HI, USA, Januar 2024
A Re-Parameterized Vision Transformer (ReVT) for Domain-Generalized Semantic Segmentation
Jan-Aike Termöhlen, Timo Bartels, and Tim Fingscheidt
In Proc. of ICCV-Workshops, pages 4376–4385, Paris, France, October 2023
Survey on Unsupervised Domain Adaptation for Semantic Segmentation for Visual Perception in Automated Driving
Manuel Schwonberg*, Joshua Niemeijer*, Jan-Aike Termöhlen*, Jörg P. Schäfer, Nico M. Schmidt, Hanno Gottschalk, and Tim Fingscheidt (* equal contribution)
IEEE Access, vol. 11, pages 54296–54336, May 2023
On the Choice of Data for Efficient Training and Validation of End-to-End Driving Models
Marvin Klingner, Konstantin Müller, Mona Mirzaie, Jasmin Breitenstein, Jan-Aike Termöhlen, and Tim Fingscheidt
In Proc. of CVPR-Workshops, pages 4803–4812, New Orleans, LA, USA, June 2022
Unsupervised BatchNorm Adaptation (UBNA): A Domain Adaptation Method for Semantic Segmentation Without Using Source Domain Representations
Marvin Klingner, Jan-Aike Termöhlen, Jacob Ritterbach, and Tim Fingscheidt
In Proc. of WACV-Workshops, pages 210–220, Waikoloa, HI, USA, Jan 2022
Reconfigurable Intelligent Surface Enabled Spatial Multiplexing with Fully Convolutional Network
Bile Peng, Jan-Aike Termöhlen, Cong Sun, Danping He, Ke Guan, Tim Fingscheidt, and Eduard A. Jorswieck
arXiv:2201.02834, pages 1–25, Januar 2022
Continual Unsupervised Domain Adaptation for Semantic Segmentation by Online Frequency Domain Style Transfer
Jan-Aike Termöhlen, Marvin Klingner, Leon J. Brettin, Nico M. Schmidt, and Tim Fingscheidt
In Proc. of ITSC, pages 2881–2888, Indianapolis, IN, USA, September 2021
Quo Vadis? Meaningful Multiple Trajectory Hypotheses Prediction in Autonomous Driving
Antonia Breuer, Quy Le Xuan, Jan-Aike Termöhlen, Silviu Homoceanu, and Tim Fingscheidt
In Proc. of ITSC, pages 637–644, Indianapolis, IN, USA, September 2021
Corner Cases for Visual Perception in Automated Driving: Some Guidance on Detection %Approaches
Jasmin Breitenstein, Jan-Aike Termöhlen, Daniel Lipinski, and Tim Fingscheidt
arXiv:2102.05897, pages 1–8, February 2021
openDD: A Large-Scale Roundabout Drone Dataset
Antonia Breuer, Jan-Aike Termöhlen, Silviu Homoceanu, and Tim Fingscheidt
In Proc. of ITSC, pages 44–49, Virtual, September 2020
Self-Supervised Monocular Depth Estimation: Solving the Dynamic Object Problem by Semantic Guidance
Marvin Klingner, Jan-Aike Termöhlen, Jonas Mikolajczyk, and Tim Fingscheidt
In Proc. of ECCV, pages 1–19, Virtual, August 2020
Terminology and Analysis of Map Deviations in Urban Domains: Towards Dependability for HD Maps in Automated Vehicles
Christopher Plachetka, Niels Maier, Jenny Fricke, Jan-Aike Termöhlen, and Tim Fingscheidt
In Proc. of IV-Workshops, pages 63–70, Las Vegas, NV, USA, June 2020
Systematization of Corner Cases for Visual Perception in Automated Driving
Jasmin Breitenstein, Jan-Aike Termöhlen, Daniel Lipinski, and Tim Fingscheidt
In Proc. of IV, pages 986–993, Las Vegas, NV, USA, June 2020
Analysis of the Effect of Various Input Representations for LSTM-Based Trajectory Prediction
Antonia Breuer, Sven Elflein, Tim Joseph, Jan-Aike Bolte, Silviu Homoceanu, and Tim Fingscheidt
In Proc. of ITSC, pages 1–8, Aucklan, New Zealand, October 2019
Unsupervised Domain Adaptation to Improve Image Segmentation Quality Both in the Source and Target Domain
Jan-Aike Bolte, Markus Kamp, Antonia Breuer, Silviu Homoceanu, Peter Schlicht, Fabian Hüger, Daneil Lipinski, and Tim Fingscheidt
In Proc. of CVPR-Workshops, pages 1404–1413, Long Beach, CA, USA, June 2019
Towards Corner Case Detection for Autonomous Driving
Jan-Aike Bolte, Andreas Bär, Daniel Lipinski, and Tim Fingscheidt
In Proc. of IV, pages 366–373, Paris,France, June 2019