Image Denoising Codes using STROLLR learning, the Matlab implementation of the paper in ICASSP2017
STROLLR2D image denoising accompanies the following publication:
"When Sparsity Meets Low-Rankness: Transform Learning With Non-Local Low-Rank Constraint for Image Restoration", IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2017. [ICASSP 2017], [PDF available], [Code]
STROLLR is an image denoising framework based on a joint adaptive patch sparse and group low-rank model learning scheme (STROLLR). The proposed scheme is capable of better representing natural images by exploiting both its local sparsity and non-local similarity. Our numerical experiments show promising performance for the proposed image denoising method compared to popular prior or state-of-the-art methods.
You can download our other software packages at: My Homepage and Transform Learning Site.
Paper
In case of use, please cite our publications:
B. Wen, Y. Li, and Y. Bresler, “When Sparsity Meets Low-Rankness: Transform Learning With Non-Local Low-Rank Constraint for Image Restoration,” in Proc. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2017.
@inproceedings{wen2017strollr2d,
title = {When sparsity meets low-rankness: Transform learning with non-local low-rank constraint for image restoration},
author = {Wen, Bihan and Li, Yanjun and Bresler, Yoram},
booktitle = {IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)},
pages = {2297--2301},
year = {2017},
organization={IEEE}
}
All codes are subject to copyright and may only be used for non-commercial research. In case of use, please cite our publication.
Contact Bihan Wen (bihan.wen.uiuc@gmail.com) for any questions.
The development of this software was supported in part by the National Science Foundation (NSF) under grants CCF 06-35234 and CCF 10-18660.