Claas Bollen

About

Claas Bollen is a researcher whose work lies at the intersection of computer vision and robotics, with a primary focus on deep learning for semantic segmentation and human perception. His most influential contribution, the 2016 paper "Deep learning for human part discovery in images," has garnered 93 citations and addresses the critical challenge of segmenting human body parts in standard RGB images. This work has direct applications in robotics, particularly in learning from demonstration and human-robot handovers, where understanding human pose and limb positioning is essential for safe and effective interaction. Bollen further advanced the field with his 2017 study "Efficient and robust deep networks for semantic segmentation" (31 citations), where he explored and introduced three novel up-convolutional network architectures designed to improve both the computational efficiency and robustness of pixel-level scene understanding. By tackling the dual challenges of accuracy and real-time performance, Bollen’s research provides foundational tools that enable robots to better perceive and interact with humans and their environments, marking him as a notable contributor to the development of more capable and responsive robotic systems.

Research Focus

Key Achievements

Top Papers

1. 1
   Deep learning for human part discovery in images

   93 citations · 2016
2. 2
   Efficient and robust deep networks for semantic segmentation

   31 citations · 2017

Key Collaborators

Contact & Links