Ben Glocker

Institute of Group Analysis, Imperial College London, NIHR Imperial Biomedical Research Centre

Papers

Total Citations

H-Index

About

Ben Glocker is a leading researcher at the intersection of robotics, computer vision, and medical imaging, with a primary focus on surgical scene understanding and autonomous navigation. His work addresses critical challenges in flexible robotic systems, particularly concentric tube robots, where he developed unified tracking and shape estimation techniques essential for safe intracorporeal control—a contribution that has garnered over 34 citations and laid groundwork for real-time surgical robotics. Glocker also made significant strides in dense visual SLAM, introducing a novel approach for real-time, globally consistent surfel-based mapping of room-scale environments using RGB-D cameras, published at Robotics: Science and Systems (2015, 27 citations). More recently, he has pioneered advanced training strategies for medical image segmentation, including paced-curriculum distillation and confidence-aware label smoothing, which enhance model robustness and uncertainty quantification in surgical scene understanding. His work on curriculum learning and self-paced training, published in 2023, demonstrates how difficulty-based sample ordering improves performance in robotic vision tasks. With a career spanning foundational robotics algorithms to cutting-edge deep learning for healthcare, Glocker’s research consistently bridges theory and application, making him a key figure in advancing safe, intelligent robotic systems for clinical use.