Learning Analysis Sparsity Priors - Archive ouverte HAL Access content directly
Conference Papers Year : 2011

Learning Analysis Sparsity Priors


This paper introduces a novel approach to learn a dictionary in a sparsity-promoting analysis-type prior. The dictionary is opti- mized in order to optimally restore a set of exemplars from their degraded noisy versions. Towards this goal, we cast our prob- lem as a bilevel programming problem for which we propose a gradient descent algorithm to reach a stationary point that might be a local minimizer. When the dictionary analysis operator specializes to a convolution, our method turns out to be a way of learning generalized total variation-type prior. Applications to 1-D signal denoising are reported and potential applicability and extensions are discusses.
Fichier principal
Vignette du fichier
PeyreFadiliSampta11.pdf (141.49 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-00542016 , version 1 (01-12-2010)


  • HAL Id : hal-00542016 , version 1


Gabriel Peyré, Jalal M. Fadili. Learning Analysis Sparsity Priors. Sampta'11, 2011, Singapour, Singapore. 4 pp. ⟨hal-00542016⟩
754 View
831 Download


Gmail Facebook X LinkedIn More