Always supportive and understanding.
This comment is not public.
Luc Van Gool is a professor in the Faculty of Engineering Science at KU Leuven, affiliated with the Department of Electrical Engineering (ESAT), where he leads the VISICS research group (VIS for Industry, Communication, and Services) within ESAT-PSI, the Processing of Speech and Images lab. He serves as a member of the KU Leuven Brain Institute and co-leads the Language, Speech, and Vision pillar at Leuven.AI. His research specializations encompass computer vision, deep neural networks, multi-task learning, object recognition, tracking, depth estimation, sensor fusion, domain adaptation, weakly-supervised learning, and applications in autonomous driving, including visual scene understanding, motion exploitation for online-adaptive scene understanding, and response to emergency services. He has been promotor or co-promotor for numerous PhD projects, such as Visual Scene Understanding for Model Selection in Autonomous Driving (2022-2026), Response of Autonomous Cars to Emergency Services (2021-2025), Exploring Unsupervised Learning for Computer Vision Tasks with Neural Networks (2018-2023), Multi-Task Learning for Visual Scene Understanding (2018-2022), and Large Scale Video Understanding (2016-2022). Collaborations include major projects with Toyota on autonomous driving and the Macchina AI initiative with colleagues Tinne Tuytelaars, Matthew Blaschko, and Marie-Francine Moens.
Van Gool holds an emeritus position at ETH Zurich, where he was full professor since 1998 and headed the Computer Vision Laboratory in the Department of Information Technology and Electrical Engineering. He received his degree in electromechanical engineering from KU Leuven in 1981. His contributions have earned awards including the PAMI Distinguished Researcher Award (2017), David Marr Prize, Koenderink Award, Helava Prize, Tsuji Award, Most Influential Scholar Honorable Mention (2021), and ranking 16th worldwide among computer scientists by h-index (2020). Key publications feature highly cited works like SURF: Speeded Up Robust Features (2008, over 22,000 citations), The PASCAL Visual Object Classes (VOC) Challenge (2010, over 27,000 citations), and recent ECCV papers such as Neural Vector Fields for Implicit Surface Representation and Inference (2025), Bayesian Self-training for Semi-supervised 3D Segmentation (2025), and Four Ways to Improve Verbo-visual Fusion for Dense 3D Visual Grounding (2025). With over 279,000 citations, his research has significantly impacted computer vision, fostering advancements in machine learning and practical applications.
