Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add NLMeans denoiser in ImageProcessing #1181

Merged
merged 5 commits into from
Dec 13, 2022
Merged

Conversation

tanstella
Copy link

@tanstella tanstella commented Jun 29, 2022

Description

This PR adds a new denoising method (Non-Local Means from OpenCV) in the ImageProcessing node.

NL Means support in Meshroom is added in alicevision/Meshroom#1719.

Implementation remarks

  • Implementation inspired by bilateralFilter
  • NLMeans is a 8-bit denoiser so conversion from float to unsigned char is managed

p12tic
p12tic previously approved these changes Oct 3, 2022
Stella TAN and others added 5 commits December 13, 2022 17:55
add associated parameters
OpenCV's core/core.hpp and photo/cuda.hpp are not needed to use the
implementation of fast non-local means denoising.
This commit corrects a  few typos in the comments dedicated to the
NL Means and it updates the default value of the "filterStrengthColor"
parameter to 10.f (instead of 5.f), as recommended by OpenCV's
documentation.
This commit moves the implementation of cvMatBGRToImageRGBA into a
dedicated template function (cvMatBGRToImageRGBAImpl). The documentation
for both methods is added and updated where needed.
This commit harmonizes the descriptions of existing and unmodified
functions, and corrects an error with the Doxygen tags.
@fcastan fcastan force-pushed the DenoiseNLMeansOpenCV branch from 68365fe to 7e029a7 Compare December 13, 2022 17:36
@fabiencastan fabiencastan merged commit b2abc45 into develop Dec 13, 2022
@fabiencastan fabiencastan deleted the DenoiseNLMeansOpenCV branch December 13, 2022 21:48
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants