Tim Broedermann
About Me
I am a computer vision researcher with a PhD from the Computer Vision Lab at ETH Zurich, supervised by Prof. Luc Van Gool.
My research focuses on multimodal fusion for robust semantic perception and 3D reconstruction, with an emphasis on all-weather scene understanding for autonomous systems. I develop adaptive sensor-fusion architectures across cameras, LiDAR, and radar, and study how to make perception robust to adverse conditions.
Before joining the lab, I completed an MSc in Robotics, Systems and Control and a BSc in Mechanical Engineering, both at ETH Zurich. I also spent a semester at the University of British Columbia and attended ICVSS 2024 in Sicily.
Research Interests
- Multimodal and multi-sensor fusion
- Robust semantic perception in adverse weather
- 3D reconstruction and neural scene representations
- Sensor calibration and cross-modal alignment
- Learning-based perception for autonomous driving
Selected Publications
- DGFusion: Depth-Guided Sensor Fusion for Robust Semantic Perception (RA-L 2026)
- PBR-NeRF: Inverse Rendering with Physics-Based Neural Fields (CVPR 2025)
- CAFuser: Condition-Aware Multimodal Fusion for Robust Semantic Perception (RA-L 2025)
- MUSES: Multi-Sensor Semantic Perception Dataset for Driving under Uncertainty (ECCV 2024)
- Radar Fields: Frequency-Space Neural Scene Representations for FMCW Radar (SIGGRAPH 2024)
- HRFuser: Multi-resolution Sensor Fusion for 2D Object Detection (ITSC 2023)
Teaching
- Head TA, Computer Vision and AI for Autonomous Cars (ETH Zurich, 2023/24)
- Head TA, Vision and Control in RoboCup (ETH Zurich, 2022)
- TA, Computer Vision and AI for Autonomous Cars (ETH Zurich, 2024/25)
- TA, Deep Learning for Autonomous Driving (ETH Zurich, 2021/22)
- TA, Werkstoffe und Fertigung I+II (ETH Zurich, 2016/17)
Contact
Email: timbroed -at- gmail -dot- com