LISA has developed expertise both in image analysis/pattern recognition and computer graphics. In the field of image analysis and pattern recognition, this unit develops new methods for object segmentation, recognition or tracking in 2D and 3D problems, multi-modal image registration, as well as statistical learning methods applied to image and data classification. Developed algorithms are related to biomedical, industrial and HMI (Human Machine Interface) applications. Image synthesis and virtual reality research activities are oriented towards medical and real time applications such as: patient home care, computer-assisted surgery, 3D real-time rendering of complex geometry and object collision, exact visibility computation and gestural interface.
Most of the aforementioned 2D/3D audio-visual signal processing algorithms require heavy number crunching on large data sets and must hence rely on efficient multi-core parallelization to ensure low-latency, real-time processing. Best performances are achieved when thoroughly trading off the intricate relationship between the application requirements, the algorithmic structure and the architecture’s multilevel memory hierarchy. In particular, General Purpose GPU programming and its efficient partitioning into regular processing kernels with minimal data dependency crossovers calls for a complementary expertise over the full application-algorithm-architecture value chain. In particular, the recent Deep Learning development for signal and image processing gain in efficacy thanks to the massively parallel computing power available on modern GPU cards; this research topic has been chosen for three new phd thesis in the lab.
LISA, following a problem-centered approach, tackles all hardware and software aspects of the chain in multidisciplinary teams (MDs, biologists, engineers, computer scientists, archaeologists, artists ...) over multi-institutional collaborations to deliver functional applications. The research is funded both by institutional/public funds and industry collaborations. LISA's achievements include one patent, several highly cited biomedical papers, implementation of acquisition and thermoregulation devices for live cell imaging, multi-media event organization and international cultural heritage projects.
Image super-resolution and image fusion
Relevant Similarities In Images through DNN
Patient-Derived Tumor Growth Modeling from Multiparametric Analysis of Combined Dynamic PET/MR Data
Platform for Imaging in Clinical Research in Brussels
Proton Therapy Research in Wallonia
Augmented RealiTy Unilateral RehabilitatiOn
COllaborative Consortium for the early detection of LIver CANcer
Automatic Recognition for Map Update by Remote Sensing
Star Image Recognition and Its Application in Guidance for Spacecrafts
Multi-modal sensor for road traffic analysis
The LISA is closely connected to two ULB's technological platforms
The general context of the project is the intelligent management of Piping and Instrumentation Diagrams (P&ID). Industry generates a huge number of schematics for various purposes. Most of the time these documents are created and modified using specialized drawing software. By the end the human operator (and therefore …
With this project, we propose to bring century-old analog seismic data and metadata compliant with modern standards by bridging two domains of expertise, namely seismology and machine learning.
The goal of the project is to develop methodologies to make pre-digital seismic (i.e. scanned seismogram images) data compliant with …