Luc Van Gool

TL;DR: An automatic annotation scheme in which a latent topic model is employed to generate topic distributions from weighted text and then modified these distributions based on visual similarity is proposed, which is able to generate accurate annotations, even for locations only seen in a single episode.

TL;DR: In this article, the layout of the different subtextures is modeled as a texture, which can be generated automatically by segmenting the texture into sub-textures and then filling in the layout with the appropriate sub-texture.

TL;DR: A novel automatic recognition framework for hand-written mensural music which takes a scanned manuscript as input and yields as output modern music scores, and works as a complete pipeline which integrates both recognition and transcription.

TL;DR: This work presents a new audio-visual corpus for possibly the two most important modalities used by humans to communicate their emotional states, namely speech and facial expression in the form of dense dynamic 3D face geometries and introduces an acquisition setup for labeling the data with very little manual effort.

TL;DR: Zhang et al. as mentioned in this paper proposed a ranking generative adversarial network (RGAN) to discover multiple object instances for three cases: synthesizing a picture of a specific object within a cluttered scene, localizing different categories in images for weakly supervised object detection, and improving object discovery in object detection pipelines.