Skip to Main content Skip to Navigation
Conference papers

Learning local regularization for variational image restoration

Abstract : In this work, we propose a framework to learn a local regularization model for solving general image restoration problems. This regularizer is defined with a fully convolutional neural network that sees the image through a receptive field corresponding to small image patches. The regularizer is then learned as a critic between unpaired distributions of clean and degraded patches using a Wasserstein generative adversarial networks based energy. This yields a regularization function that can be incorporated in any image restoration problem. The efficiency of the framework is finally shown on denoising and deblurring applications.
Complete list of metadata

https://hal.archives-ouvertes.fr/hal-03139784
Contributor : Jean Prost Connect in order to contact the contributor
Submitted on : Friday, February 12, 2021 - 12:12:51 PM
Last modification on : Tuesday, March 8, 2022 - 9:26:01 AM
Long-term archiving on: : Thursday, May 13, 2021 - 6:43:36 PM

File

learning_local_regularization....
Files produced by the author(s)

Identifiers

  • HAL Id : hal-03139784, version 1
  • ARXIV : 2102.06155

Collections

CNRS | ANR

Citation

Jean Prost, Antoine Houdard, Andrés Almansa, Nicolas Papadakis. Learning local regularization for variational image restoration. International Conference on Scale Space and Variational Methods in Computer Vision (SSVM'21), May 2021, Cabourg, France. pp.358-370. ⟨hal-03139784⟩

Share

Metrics

Record views

99

Files downloads

41