We propose a convolutional recurrent sparse auto-encoder model. The model
consists of a sparse encoder, which is a convolutional extension of the learned
ISTA (LISTA) method, and a linear convolutional decoder. Our strategy offers a
simple method for learning a task-driven sparse convolutional dictionary (CD),
and producing an approximate convolutional sparse code (CSC) over the learned
dictionary. We trained the model to minimize reconstruction loss via gradient
decent with back-propagation and have achieved competitive results to KSVD
image denoising and to leading CSC methods in image inpainting requiring only a
small fraction of their run-time.
%0 Journal Article
%1 sreter2017learned
%A Sreter, Hillel
%A Giryes, Raja
%D 2017
%K compression sparsity
%T Learned Convolutional Sparse Coding
%U http://arxiv.org/abs/1711.00328
%X We propose a convolutional recurrent sparse auto-encoder model. The model
consists of a sparse encoder, which is a convolutional extension of the learned
ISTA (LISTA) method, and a linear convolutional decoder. Our strategy offers a
simple method for learning a task-driven sparse convolutional dictionary (CD),
and producing an approximate convolutional sparse code (CSC) over the learned
dictionary. We trained the model to minimize reconstruction loss via gradient
decent with back-propagation and have achieved competitive results to KSVD
image denoising and to leading CSC methods in image inpainting requiring only a
small fraction of their run-time.
@article{sreter2017learned,
abstract = {We propose a convolutional recurrent sparse auto-encoder model. The model
consists of a sparse encoder, which is a convolutional extension of the learned
ISTA (LISTA) method, and a linear convolutional decoder. Our strategy offers a
simple method for learning a task-driven sparse convolutional dictionary (CD),
and producing an approximate convolutional sparse code (CSC) over the learned
dictionary. We trained the model to minimize reconstruction loss via gradient
decent with back-propagation and have achieved competitive results to KSVD
image denoising and to leading CSC methods in image inpainting requiring only a
small fraction of their run-time.},
added-at = {2019-06-02T18:08:55.000+0200},
author = {Sreter, Hillel and Giryes, Raja},
biburl = {https://www.bibsonomy.org/bibtex/2e080d66c9ff7b97f0eb2b3b2ef565832/kirk86},
description = {[1711.00328] Learned Convolutional Sparse Coding},
interhash = {0320482cf6fc76b3d96a077e04fe4eb6},
intrahash = {e080d66c9ff7b97f0eb2b3b2ef565832},
keywords = {compression sparsity},
note = {cite arxiv:1711.00328},
timestamp = {2019-06-02T18:08:55.000+0200},
title = {Learned Convolutional Sparse Coding},
url = {http://arxiv.org/abs/1711.00328},
year = 2017
}