Skip to content

Improving color consistency across multiple images via global optimization on the color remapping curves. [ICCVW 2017 & ISPRS Journal 2019]

Notifications You must be signed in to change notification settings

MenghanXia/ColorConsistency

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

45 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Introduction

This project is developed for optimizing the color consistency across images and guarantee the imaging quality of individual image meanwhile. we parameterize the color remapping curve as transform model, and express the constraints of color consistency, contrast and gradient in an uniform energy function. It can be formulated as a convex quadratic programming problem which provides the global optimal solution efficiently.

This C++ implemented algorithm is described in:
"Color Consistency Correction Based on Remapping Optimization for Image Stitching", ICCV Workshop 2017.
"A Closed-Form Solution for Multi-View Color Correction with Gradient Preservation", ISPRS Journal 2019 (Extended version).
This program is free for personal, non-profit and academic use. If you have any question, please contact: menghanxyz@gmail.com (Menghan Xia)

Here is an example for demonstration below:

Usage

1. Dependent Libarary [compulsory]:

OpenCV 2.4.9 is recommended.

2. Project Configure:

This procedure is developed on Visual Studio 2010 under Windows 8.1 system, where the source code is organized with CMakeLists. So, before opening it in Visual Studio, you need to configure the project with the software named CMake.

3. Running and Test:

There will be a "Data" folder in the decompressed files, where three created folders exist:

  • "Cache" : creat a text file that describes the adjacent relationships of each image. <example contained>
  • "Images" : input your source images that were aligned via image stitching algorithms (e.g., PanoramaStudio or AutoStitching). <example contained>
  • "Results": automatically save the processed results.

Besides, to use the existing "Data" directory successfully, do not forget to UPDATE the path variable 'baseDir' as the absolute path of your directory "Data" in the source file CoColour/Source/Utils/util.h [line 16]

So far, you can run the procedure and see the color correction results now.

Citation

If any part of our paper and code is helpful to your work, please generously cite with:

@inproceedings{XiaYXZX17,
  author    = {Menghan Xia and Jian Yao and Renping Xie and Mi Zhang and Jinsheng Xiao},
  title     = {Color Consistency Correction Based on Remapping Optimization for Image Stitching},
  booktitle = {{IEEE} International Conference on Computer Vision Workshops (ICCVW)},
  year      = {2017}
}

About

Improving color consistency across multiple images via global optimization on the color remapping curves. [ICCVW 2017 & ISPRS Journal 2019]

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published