Télécom ParisTech, LTCI, Paris, FR

In 2010, Joseph Salmon finished his PHD in Statistics under the supervision of Dominique Picard and Erwan Le Pennec at the Laboratoire de Probabilités et de Modélisation Aléatoire (LPMA ) in Université Paris Diderot.
His thesis is about the denoising of digital images combining patch-based method and statistical aggregation. Salmon is especially interested in topics as: aggregation of estimators, Patch based denoising methods, Non-parametric regression and On-line learning.

From 2011 to the end of 2012, Joseph Salmon was a Post Doctoral Associate at Duke university working with Rebecca Willett in the Network and Imaging Science Laboratory (NISLab) . During this period, he also started lecturing and is sharing his knowledge since 2013 at TELECOM ParisTech.



  • Mandatory Critical Points of 2D Uncertain Scalar Fields,
    D. Guenther, J. Salmon, J. Tierny, Computer Graphics Forum, aug 2014
  • Reconstruction Stable par Régularisation Sécomposable Analyse.,
    J. Fadili, G. Peyré, S. Vaiter, Ch.-a. Deledalle, J. Salmon, GRETSI, sep 2013
  • Stable Recovery with Analysis Decomposable Priors,
    J. Fadili, G. Peyré, S. Vaiter, Ch.-a. Deledalle, J. Salmon, SampTA, jul 2013
  • Learning Heteroscedastic Models By Convex Programming Under Group Sparsity ,
    A. Dalalyan, M. Hebiri, K. Meziani, J. Salmon, ICML, jun 2013


More info: josephsalmon.eu

Speaker at
8-10 October 2014, Leuven, Belgium