Approximate Bayesian Computation, stochastic algorithms and non-local means for complex noise models - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2014

Approximate Bayesian Computation, stochastic algorithms and non-local means for complex noise models

Abstract

In this paper, we present a stochastic NL-means-based denoising algorithm for generalized non-parametric noise models. First, we provide a statistical interpretation to current patch-based neighborhood filters and justify the Bayesian inference that needs to explicitly accounts for discrepancies between the model and the data. Furthermore, we investigate the Approximate Bayesian Computation (ABC) rejection method combined with density learning techniques for handling situations where the posterior is intractable or too prohibitive to calculate. We demonstrate our stochastic Gamma NL-means (SGNL) on real images corrupted by non-Gaussian noise.
Fichier principal
Vignette du fichier
2014-ICIP-GNL.pdf (1.34 Mo) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-01103322 , version 1 (23-01-2015)

Identifiers

  • HAL Id : hal-01103322 , version 1

Cite

Charles Kervrann, Philippe Roudot, François Waharte. Approximate Bayesian Computation, stochastic algorithms and non-local means for complex noise models. IEEE International Conference on Image Processing, Oct 2014, Paris, France. pp.4. ⟨hal-01103322⟩
229 View
432 Download

Share

Gmail Facebook Twitter LinkedIn More