Research Topics

Associated with Inria Saclay, the Centre de Vision Numérique, created in 2011, is at the intersection between mathematics and computer science. It is in search of mathematical models and their IT solutions for automatic structuring, interpreting and understanding big (visual) data with a focus on machine learning, computer vision and discrete models in biomedical image analysis.

Computer Vision

Image reconstruction, border detection, segmentation with or without modeling, flow estimation and monitoring, image analysis, object recognition and large-scale 3D modeling based on a grammar

Machine Learning and Optimization

Self-learning, probabilistic graphical models, multiple instance learning, structured output regression, kernel methods, and multitask, online, or transfer learning, etc.

Biomedical Image Analysis (GALEN-Inria team)

Compressed reconstruction and detection, tumor detection, organ segmentation, image registration and deformable fusion, longitudinal modeling of organs, virtual anatomy, population studies and understanding of the brain, etc.

The Center of Visual Computing of CentraleSupelec & Inria, Saclay, Ile-de-France, organized a summer school in Biomedical Image Analysis: Modalities, Methodologies & Clinical Research at the Institut Henri Poincaré at the heart of Paris. This was an official event of the Medical Image Computing and Computer Assisted Intervention Society (MICCAI). Please find out all the lectures given for this special event.

Fields of Application

  • Complex industrial systems (automation, optical sorting, robotics, control systems, non-destructive testing)

  • Automotive Industry (assisted driving, pedestrian detection, cruise control, parking assistance)

  • Health (computer-aided diagnostics, multimodal sensors, data mining, biomarker imaging, computer-aided surgery) 

 

Key Figures

  • Instructor-researchers and researchers: 5
  • PhD students: 17
  • Technical and administrative staff: 3
  • Interns: 8
  • Patents: 1
  • Publications: 22

 

Academic Partners

Inria (France), École des Ponts ParisTech (France), Henri Mondor University Hospital (France), Georges Pompidou European Hospital (France), Pitié-Salpêtrière Hospital (France), Montpelier University Hospital (France), CentraleSupélec (France), Stanford University (USA), StonyBrook University (USA), University of Pennsylvania (USA), UCLA (USA), Technical University of Munich (Germany), University of Lugano (Switzerland), University of Oxford (UK), University College London (UK), University of Oulu (Finland), École Polytechnique de Montréal (Canada), International Institute of Information Technology, Hyderabad (India)

 

Business and Research Clusters

Digiteo, Medicen, Cap Digital

 

Industrial Partners

GE Healthcare, Siemens Medical Solutions, Intrasense, LLTech
 

Contact

 

Website: http://cvn.ecp.fr/

Director: Jean-Christophe PESQUET
Tel.: +33 (0)1 41 13 17 85
Fax: +33 (0)1 41 13 10 06

Email: Jean-Christophe.pesquet@centralesupelec.fr 

 

Latest submissions

Communication on a congress
04/13/2021
3D Unsupervised Kidney Graft Segmentation Based on Deep Learning and Multi-Sequence MRI
Leo Milecki, Sylvain Bodard, Jean-Michel Correas, Marc-Olivier Timsit, Maria Vakalopoulou
Communication on a congress
04/01/2021
Contrast-enhanced brain MRI synthesis with deep learning: key input modalities and asymptotic performance
Alexandre Bône, Samy Ammari, Jean-Philippe Lamarque, Mickael Elhaik, Émilie Chouzenoux, François Nicolas, Philippe Robert, Corinne Balleyguier, Nathalie Lassau, Marc-Michel Rohé
Communication on a congress
04/01/2021
A Deep Learning Approach for Improved Segmentation of Lesions Related to Covid-19 Chest CT Scans
Vlad Vasilescu, Ana Neacsu, Emilie Chouzenoux, Jean-Christophe Pesquet, Corneliu Burileanu
Pre-submission / Working document
03/02/2021
Optimized Population Monte Carlo
Víctor Elvira, Emilie Chouzenoux
Pre-submission / Working document
02/18/2021
Optimizing persistent homology based functions
Mathieu Carriere, Frédéric Chazal, Marc Glisse, Yuichi Ike, Hariprasad Kannan
Browse all laboratory submissions on HAL