Alternative design of DeepPDNet in the context of image restoration
Image and Video Processing
2022-02-22 v1 Computer Vision and Pattern Recognition
Machine Learning
Abstract
This work designs an image restoration deep network relying on unfolded Chambolle-Pock primal-dual iterations. Each layer of our network is built from Chambolle-Pock iterations when specified for minimizing a sum of a -norm data-term and an analysis sparse prior. The parameters of our network are the step-sizes of the Chambolle-Pock scheme and the linear operator involved in sparsity-based penalization, including implicitly the regularization parameter. A backpropagation procedure is fully described. Preliminary experiments illustrate the good behavior of such a deep primal-dual network in the context of image restoration on BSD68 database.
Keywords
Cite
@article{arxiv.2202.09810,
title = {Alternative design of DeepPDNet in the context of image restoration},
author = {Mingyuan Jiu and Nelly Pustelnik},
journal= {arXiv preprint arXiv:2202.09810},
year = {2022}
}