Projects
3D semantic change detection exploiting multi-modal Earth Observation big data KU Leuven
Develop robust and efficient machine learning and deep learning methods to interpret geolocated imagery and process the pixel information into useful information for construction stakeholders
3D image processing and fusion on multi-modality sensing data for autonomous driving KU Leuven
With the growth of sensing devices such as RGB cameras and LiDAR sensors, 3D image processing on multi-modality sensing data serves as an essential foundation for autonomous driving. These sensing devices and corresponding computer vision algorithms are necessary for a self-driving car to observe the surrounding environment and act accordingly. By virtue of the recent development of neural networks, sensing technologies, such as denoising, ...
Cross-Modal and Cross-Lingual Information Search and Retrieval. KU Leuven
This project focuses on extending and combining the multi-idiomatic and multi-modal settings to highly noisy and unstructured data, as naturally found on the Web (e.g., on social media, e-commerce sites, blog posts, etc.). I propose to develop novel multi-modal, multi-lingual and multi-idomatic retrieval and search models, grounded on the successful deep learning framework, to create innovative technology that jointly performs text analysis ...
Construction Digital Twins for automated digital construction site monitoring KU Leuven
The construction industry must lower failure costs that are caused by delays and construction errors. To this end, the progress and quality of construction sites must be systematically and digitally monitored and the designs updated to as-built conditions.
The objective of this project is to fulfil the above needs by conceiving construction digital twins –a novel BIM that models the life-cycle of the construction’s execution phase ...
Exploring Unsupervised Learning for Computer Vision Tasks with Neural Networks KU Leuven
Traditional supervised learning algorithms for computer vision tasks usually rely on large extensively annotated datasets. However, this labeling process can be expensive, biased, and susceptible to ambiguity.
The thesis explores self-supervised or unsupervised learning as a viable alternative to overcome these obstacles.
Specifically, it tackles fundamental perception tasks via neural networks and aims to automatically discover ...
Artificial Language Understanding in Robots (ATLANTIS) Vrije Universiteit Brussel
Expressive Music Interaction Ghent University
This project aims at finding solutions for the semantic gap between music experience and digital encoding of music. To close this gap, this project aims at investigating how the human body can be given a technological extension so taht fluent activities in music production and music information retrieval become possible. The focus is on subjective-oriente approaches (such as multi-modality, cultural/social contexts, subjective factors) that ...
New Language Constructs and Inferences for the Knowledge Base Paradigm: A Business and Multi-agent Perspective KU Leuven
In Artificial Intelligence, the scientific field of Knowledge Representation and Reasoning (KRR) is concerned with developing formal languages, to represent knowledge, and inference methods to solve tasks using that knowledge. Most of the existing approaches develop a formal language (a logic) together with an inference, to solve some type of computational task. The recently proposed Knowledge Base (KB) paradigm applies a strict separation of ...