Projects
Computer simulation of material deposition and fuel retention in castellated structures of fusion machines Ghent University
To ensure thermo-mechanical stability and operational durability of the plasma facing components of ITER, it is foreseen to split them into small-size cells. The introduction of such castellated structures raises serious concerns about impurity accumulation and fuel retention in the gaps between the cells that may lead to unacceptable levels of fuel inventory (radioactive tritium). Modelling will be performed to understanding the basic ...
Using Generated Data to Train Computer Vision Models. Hasselt University
Improving PIPAC therapy responses in cancer patients with peritoneal metastases using robust computer vision Ghent University
Peritoneal metastasis (PM) occurs in advanced stages of ovarian and gastro-intestinal cancers. Patients with PM have a poor prognosis and their quality of life is severely compromised. Pressurized intraperitoneal aerosol chemotherapy (PIPAC) is a promising treatment option but responses are hard to predict. Indeed, standard clinical, microscopic, and medical imaging modalities are currently limited in their potential to quantify PM and ...
Remote in-line Vision system for monitoring of the Damage Progression (ReViDaP-2.0) in high power wind turbine gearboxes Ghent University
Increasing offshore wind energy capacity together with a demand for maximum reliability raises the need for structure and machine maintenance based on (semi-)automated inspections. Smart maintenance of mechanical constructions1 does already exist but machine maintenance is behind. Recently, Soete Laboratory developed a toolkit for labscale machine monitoring as a proof of concept within the IOF StarTT-project ReViDaP-1.0 ...
A modular consistent, disciminative framework for structured output learning in computer vision. KU Leuven
State of the art computer vision systems are fundamentally reliant on statistical learning to optimize performance on a specific application. Currently, statistical frameworks in computer vision are typically based on classification and regression, probabilistic graphical models, or discriminative structured prediction frameworks such as the Structured Output Support Vector Machine (SOSVM). Although some of the best performing ...
A Modular, Consistent, Discriminative Framework for Structured Output Learning in Computer Vision. KU Leuven
State of the art computer vision systems are fundamentally reliant on statistical learning to optimize performance on a specific application. Currently, statistical frameworks in computer vision are typically based on classification and regression, probabilistic graphical models, or discriminative structured prediction frameworks such as the Structured Output Support Vector Machine (SOSVM). Although some of the best performing computer vision ...
Weakly supervised machine learning algorithms for object recognition in-the-wild and entity linking in videos KU Leuven
With the proliferation of video-rich data on the Internet, there is a pressing need for search tools that can retrieve not only relevant videos from a corpus, but also relevant snippets within a video. For retrieving relevant videos, current search technologies hinge on labor-intensive manual annotation of tags, which are subjective and often incomplete. To fully automate search and retrieval systems, we need tools that can understand the ...
Machine Learning Acceleration on Heterogeneous Platforms KU Leuven
Advances in the fields of biomedical sensors, wearables and medical implants, in combination with state-of-the-art algorithms from signal processing, machine learning and artificial intelligence, are transforming the healthcare landscape. Systems built around these technologies enable remote health monitoring, improve patient care, detect life-threatening conditions or even predict health events. Yet the full integration of such technologies ...