Hyperspectral Single-pixel Image Reconstruction using Nonnegative Matrix Factorization and Deep Learning
Recrutement en cours/passé: 
Recrutement en cours
jeremy.cohen@cnrs.fr nicolas.ducros@creatis.insa-lyon.fr

The CREATIS laboratory announce the opening of a six-month internship position, starting in March 2023. An ANR-funded PhD position will be opened in October 2023 to continue on the same topic.

The intern will conduct his work at the intersection of signal processing, numerical optimization, deep learning and optics. The goal of this internship is to develop a reconstruction algorithm for specral images acquired by a single-pixel camera, so as to allow optical biopsy during brain surgery.

 The skills acquired during the internship are the following:

  • Programming in Python (numpy/pytorch in particular), collaborative development (git and github)
  • Convex and non-convex numerical optimization (projected/proximal gradient descent)
  • Linear algebra and inverse problems (ill-posed problems, regularization)
  • Deep learning (neural network design and optimization, automatic differentiation)
  • Hyperspectral imaging

More information available in the pdf linked on the left side panel.

Contact info: