Image denoising using learned overcomplete representations

Phil Sallee, Bruno A. OlshausenICIP 2003 · OpenReview

Abstract

We describe a method for learning sparse multiscale image representations using a sparse prior distribution over the basis function coefficients. The prior consists of a mixture of a Gaussian and a Dirac delta function, and thus encourages coefficients to have exact zero values. Coefficients for an image are computed by sampling from the resulting posterior distribution with a Gibbs sampler. Denoising using the learned image model is demonstrated for some standard test images, with results that compare favorably with other denoising methods.
BibTeX
@inproceedings{sallee2003image,
  author = {Phil Sallee and Bruno A. Olshausen},
  title = {{Image denoising using learned overcomplete representations}},
  booktitle = {ICIP},
  year = {2003},
  url = {https://openreview.net/forum?id=xNLO_JW42yff},
}

Loading related content...

Figure