Video Denoising with Optical Flow Estimation

• article
• demo

BibTeX info

Communicated by Marc Lebrun
Demo edited by Jose-Luis Lisani

Abstract

In this paper we describe the implementation of state-of-the-art video denoising algorithm SPTWO [A. Buades, J.L. Lisani, M. Miladinovic, Patch Based Video Denoising with Optical Flow Estimation, IEEE Transactions on Image Processing 25 (6), 2573--2586]. This algorithm, inspired by image fusion techniques, uses motion compensation by regularized optical flow methods, which permits robust patch comparison in spatiotemporal volumes. Groups of similar patches are denoised using Principal Component Analysis, which ensures the correct preservation of fine texture and details.

Download

• full text manuscript: PDF low-res. (7.9MB) PDF (15.6MB) ^[?]
• source code: TGZ SWHID info
  Software Heritage Archive @softwareversion{sw-ipol.2018.224, title = {{Video Denoising with Optical Flow Estimation}}, author = {Antoni Buades, Jose-Luis Lisani}, date = {2018-03-27}, license = {AGPL-3.0-or-later}, version = {1.0}, swhid = {swh:1:dir:1c12a2b3c9670adf05f0160ce6791fb8c30c5558;origin=https://doi.org/10.5201/ipol.2018.224;visit=swh:1:snp:1a4ac072f7c1638f1c09190ed17ff361cb4c32dc;anchor=swh:1:rev:8e84772bf35c18609068e899cf1fe2de161c0f50}, related = {ipol.2018.224}, relatedtype = "Reference article: https://doi.org/10.5201/ipol.2018.224" } % the biblatex-software style extension is recommended for software bibitems
  Copy to clipboard

History

• Note from the editor: the source code was updated on September 30, 2019 to fix a bug that affects grayscale images. Line 41 of file libNLPCA.cpp has changed. Also file addGaussianNoise.cpp has changed so grayscale images having 3 identical channels are converted to 1 channel images. The original version is available here.
• Note from the editor: the manuscript of the article was modified on 2022-01-01 to include information about its editors. The original version of the manuscript is available here.