Learned Convolutional Sparse Coding

Hillel Sreter, Raja Giryes
11/1/2017
eess.SP

Official code: benMen87/ApproximateConvolutionalSparseCoding (44 stars, last commit 2025-12-24).

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.

View on arXivView PDF

Code Implementations(1)

benMen87/ApproximateConvolutionalSparseCodingOfficial100%

An implementation of approximate convolutional sparse coding (CSC) based on paper: https://arxiv.org/abs/1711.00328

4410PythonMay 22, 201810 months ago

Similar Papers

Closed-Loop Transcription via Convolutional Sparse Coding

86% match2023

An Efficient Approximate Method for Online Convolutional Dictionary Learning

82% match2023

Flyweight FLIM Networks for Salient Object Detection in Biomedical Images

81% match2025

Supervised sparse auto-encoders as unconstrained feature models for semantic composition

81% match2026

Inference and learning in sparse autoencoders as natural gradient flow

81% match2026

Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising

81% match1 repo1.7k2016

Cite this paper

@article{sreter2017learned,
  title  = {Learned Convolutional Sparse Coding},
  author = {Hillel Sreter and Raja Giryes},
  year   = {2017},
  eprint = {1711.00328},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/1711.00328}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/1711.00328.svg)](https://codeofpaper.com/paper/1711.00328)

Discussion