Gilles Puy

Researcher at Technicolor


Google Scholar profile

Short Bio

I am a researcher at Technicolor R&D France. Before, I was at INRIA in the PANAMA Research team
I obtained the Ph.D. degree in Electrical Engineering from Ecole Polytechnique Fédérale de Lausanne, Switzerland, in January 2014. I also obtained the M.Sc. degree in Electrical and Electronics Engineering from EPFL and the Engineering Diploma (M.Sc.) from Supélec, France, in 2009.

I received the EPFL Chorafas Foundation award
in 2013 and the EPFL Doctorate award in 2015.

My research interests are in the field of signal processing, image processing and machine learning. I work on the design of efficient sampling methods for high dimensional data (compressive sampling), low-dimensional representations, and non-linear reconstruction methods using convex and non-convex optimization. My work concerns both theory and applications, mainly in imaging problems. My complete list of publications is available here.


  • [February 2017] Our paper on compressed sensing by spread spectrum received the 2017 EURASIP Best Paper Award for the Journal on Advances in Signal Processing (paper).
  • [January 2017] Our paper on recipes for stable linear embeddings from Hilbert spaces to R^m accepted for publication at IEEE Transactions on Information Theory (paper).
  • [December 2016] Our paper on Informed source separation via compressive graph sampling accepted for an oral presentation at ICASSP (paper).
  • [May 2016] Our paper on random sampling for bandlimited graph signals accepted for publication at ACHA (paper, code).
  • [April 2016] Our paper on compressive spectral clustering accepted for a presentation at ICML (paper, code).
  • [February 2016] Starting to work as a researcher at Technicolor.