Luc Van Gool

TL;DR: In this article, a hybrid GPU-CPU approach is proposed for real-time semantic segmentation of high-definition videos using an efficient video segmentation pipeline that combines a very fast optical flow method, that is used to exploit the temporal aspect of the video and propagate semantic information from one frame to the next.

TL;DR: This work has developed an interactive visual navigation environment based on Google Street View, and introduces a soft dual attention mechanism defined over the segmented language instructions to jointly extract two partial instructions for matching the next upcoming visual landmark and the local directions to the next landmark.

TL;DR: A novel method to automatically segment the input stream and to classify the resulting segments using prior knowledge and hidden Markov models (HMMs) combined through a genetic algorithm is proposed, appropriate for general-purpose time-series classification.

TL;DR: In this article, the authors formulate the map construction and self-localization problems as convex quadratic and second-order cone programs, respectively, and apply them to indoor and outdoor datasets.

TL;DR: This paper investigates into the model-based reconstruction of complex polyhedral building roofs by using a set of 3D line segments obtained from multiview correspondence analysis of high resolution colour imagery, and chooses the optimal patch and plane configuration non-deterministically.