Deep Learning for Stance Detection: A Review and Comparison of the State-of-the-Art Approaches

Stance detection is a Natural Language Processing task that can detect if the input text is in favour, against or neutral towards a target. Research on stance detection has been growing and evolving over the last decade. In this paper, the current approaches for stance detection are discussed with a focus on the deep learning approaches. The...

Generative RGB-D Face Completion for Head-Mounted Display Removal

Virtual reality (VR) creates an exceptional experience in which users can explore virtual environments. Wearing a head-mounted display (HMD), users are able to observe a virtual world that is rendered based on their physical movement and actions. A common solution for capturing the visual and geometric information needed for the construction of...

Stylization of shadows in ski-panorama maps: A case study on Atelier Novat

Panorama maps can be defined as aerial-view paintings of geologically complex landscapes that are represented in a convenient manner to non-expert viewers. This thesis focuses on a specific variant of panorama maps known as "ski-panorama maps" or "piste-maps" drawn by the french landscape and panorama artist, Pierre Novat. These maps usually...

Evaluating Methods for Improving Crowdsourced Annotations of Images Containing Lego Bricks

Data collection by means of crowdsourcing can be costly or produce inaccurate results. Methods have been proposed for solving these problems. However, it remains unclear what methods work best in scenarios with multiple similar objects of interest present in the same image, which is important for training computer vision with applications such...

Data Generation Methods for Multi-Label Images of LEGO Bricks

Data collection and annotation have proven to be a bottleneck for computer vision applications. When faced with the task of data creation, alternative methods to traditional data collection should be considered, as time and cost may be reduced signif- icantly. We introduce three novel datasets for multi- label classification purposes on LEGO...

Faster R-CNN as an Application for Object Detection of Scattered LEGO Pieces

The benchmarks for the accuracy of the best performing object detectors to date are usually based on homogeneous datasets, including objects such as vehicles, people, animals and foods. This excludes a whole set of scenarios containing small, cluttered and rotated objects. This paper selects a state-of-the-art object detection model, Faster R...

The Current State of the Art in Multi-Label Image Classification Applied on LEGO Bricks

This paper shows how the current state of the art in image classification performs on LEGO bricks. Currently the standard image classification models with deep learning are single label image classifiers. In this paper we will convert them to work on multi-label images and subsequently evaluate how well they perform. We show how well the...

A visual analysis framework for dinghy sailing: Towards leveraging recorded training sessions

Nowadays video plays an important role in the coaching of athletes across many different sports. To make more use of the advantages videos can provide for coaching, the Dutch Sailing team is shifting from manually recording short videos towards continuously recording training sessions. This new recording approach provides opportunities and...

Lossy Geometry Compression for High Resolution Voxel Scenes

Representing geometry data as voxels allows for a massive amount of detail that can be rendered in real-time. Storing this type of data as a directed acyclic graph (DAG) has recently led to immense improvements in memory consumption, which is one of the main limitations often associated with voxel-based approaches. We present a method for...

Adaptive Multi-view Screen-space Ambient Obscurance

Screen-space ambient obscurance techniques provide a cheap, yet visually pleasing approximation to the expensive global illumination effect. However, view-dependent artifacts frequently occur due to the limited scene information available in a single view, which forms the basis of these solutions. Recent work aims at suppressing artifacts with...

BobRossNet: Accelerating rendering by partial inpainting

Before you lies the report that resulted from the Individual Research Project of Vera Hoveling. For this project, a neural network has been developed to complete partially rendered images: BobRossNet. In this document you will find the details of the project, the architecture of BobRossNet, its training, results and a reflection on its future.