L
Luc Van Gool
Researcher at Katholieke Universiteit Leuven
Publications - 1458
Citations - 137230
Luc Van Gool is an academic researcher from Katholieke Universiteit Leuven. The author has contributed to research in topics: Computer science & Segmentation. The author has an hindex of 133, co-authored 1307 publications receiving 107743 citations. Previous affiliations of Luc Van Gool include Microsoft & ETH Zurich.
Papers
More filters
Proceedings ArticleDOI
Learning a Curve Guardian for Motorcycles
TL;DR: The proposed new type of road curvature warning system for motorcycles is able to predict more accurate and safer curve trajectories, and consequently warn and improve the safety for motorcyclists.
Posted Content
Natural Illumination from Multiple Materials Using Deep Learning
Stamatios Georgoulis,Konstantinos Rematas,Tobias Ritschel,Mario Fritz,Tinne Tuytelaars,Luc Van Gool +5 more
TL;DR: Qualitative and quantitative evidence shows how both multi-material and using a background are essential to improve illumination estimations.
Posted Content
Transferring Object-Scene Convolutional Neural Networks for Event Recognition in Still Images.
TL;DR: This paper addresses the problem of event recognition by proposing a convolutional neural network that exploits knowledge of objects and scenes for event classification (OS2E-CNN), design a multi-ratio and multi-scale cropping strategy, and propose an end-to-end event recognition pipeline.
Book ChapterDOI
Image-based 3D modeling: modeling from reality
Luc Van Gool,Filip Defoort,Johannes Hug,Gregor A. Kalberer,Reinhard Koch,Danny Martens,Marc Proesmans,Maarten Vergauwen,Alexey Zalesny,Marc Pollefeys +9 more
TL;DR: A system for visits to a virtual, 3D archeological site, with a virtual guide as companion, which can ask questions using natural, fluent speech and the guide will respond and will bring the visitor to the desired place.
Posted Content
Warp Consistency for Unsupervised Learning of Dense Correspondences
TL;DR: Warp Consistency as discussed by the authors constructs an image triplet by applying a randomly sampled warp to one of the original images, and derive flow-consistency constraints arising between the triplet.