Daan de Geus

Daan de Geus serves as Assistant Professor in the Department of Electrical Engineering at Eindhoven University of Technology (TU/e), affiliated with the Mobile Perception Systems Lab. His research directly contributes to UN Sustainable Development Goals through advanced computer vision systems for sustainable mobility and urban environments.

His academic credentials include:

• PhD in Electrical Engineering (2024) from TU/e with dissertation "Advances in Scene Understanding: Towards Efficient Image Segmentation at Multiple Abstraction Levels"
• Master's degree (2018) from TU/e focused on panoptic segmentation architectures

De Geus specializes in image segmentation across multiple abstraction levels, with particular expertise in vision transformers and unsupervised domain adaptation. His research develops efficient neural network architectures that maintain accuracy while reducing computational demands, enabling real-time perception systems for mobile applications. Current work integrates diffusion models with segmentation pipelines to enhance image synthesis capabilities.

Recent publications (2024-2025) reveal a strategic shift toward foundation models and efficiency optimization. His team's ALGM framework demonstrates novel token merging techniques for vision transformers, while diffusion model research challenges conventional fine-tuning assumptions. These efforts culminated in leadership roles for the BRAVO Semantic Segmentation Challenge, establishing new benchmarks in the field.

As principal investigator for the PPS TKI NXP Smart Mobility project (2018-2023), de Geus secured industry-academic funding exceeding €500,000. His supervision portfolio includes two graduate students, with ongoing mentorship in the Mobile Perception Systems Lab where he develops next-generation perception algorithms for autonomous systems.

The Mobile Perception Systems Lab under his guidance focuses on real-time computer vision for smart mobility applications, with current projects spanning automotive perception, drone navigation, and urban infrastructure monitoring using multi-sensor fusion approaches.