Rethinking the CSC Model for Natural Images

09/12/2019
by   Dror Simon, et al.
0

Sparse representation with respect to an overcomplete dictionary is often used when regularizing inverse problems in signal and image processing. In recent years, the Convolutional Sparse Coding (CSC) model, in which the dictionary consists of shift-invariant filters, has gained renewed interest. While this model has been successfully used in some image processing problems, it still falls behind traditional patch-based methods on simple tasks such as denoising. In this work we provide new insights regarding the CSC model and its capability to represent natural images, and suggest a Bayesian connection between this model and its patch-based ancestor. Armed with these observations, we suggest a novel feed-forward network that follows an MMSE approximation process to the CSC model, using strided convolutions. The performance of this supervised architecture is shown to be on par with state of the art methods while using much fewer parameters.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
05/14/2014

Group-based Sparse Representation for Image Restoration

Traditional patch-based sparse representation modeling of natural images...
research
12/06/2019

Sparse and redundant signal representations for x-ray computed tomography

Image models are central to all image processing tasks. The great advanc...
research
06/28/2015

Patch-Based Low-Rank Minimization for Image Denoising

Patch-based sparse representation and low-rank approximation for image p...
research
06/12/2018

Fast Rotational Sparse Coding

We propose an algorithm for rotational sparse coding along with an effic...
research
05/09/2017

Convolutional Dictionary Learning via Local Processing

Convolutional Sparse Coding (CSC) is an increasingly popular model in th...
research
10/02/2018

Deep Decoder: Concise Image Representations from Untrained Non-convolutional Networks

Deep neural networks, in particular convolutional neural networks, have ...
research
04/23/2022

Gabor is Enough: Interpretable Deep Denoising with a Gabor Synthesis Dictionary Prior

Image processing neural networks, natural and artificial, have a long hi...

Please sign up or login with your details

Forgot password? Click here to reset