SPDecolor

February 25, 2020 ยท View on GitHub

Semi-parametric decolorization with Laplacian-based perceptual quality metric

% The Code is created based on the method described in the following paper:
% Q. Liu, X. Liu, Y. Wang, H. Leung. Semi-parametric decolorization with Laplacian-based perceptual quality metric,
% IEEE Trans. Circuits Syst. Video Technol., 27(9): 1856-1868, 2017.
% Version : 1.0
% The code and the algorithm are for non-comercial use only.
% Copyright 2017, Department of Electronic Information Engineering, Nanchang University.
% The current version is not optimized.

% SPDecolor - contrast preserving color to gray by Semi-Parametric optimization
% S = SPDecolor(Im, sigma)
% @Im : Input image (double), only color images are acceptable.
% @sigma : Controlling parameter defined in [1]. 5e-3 by default.
%
% Example
% ==========
% Im = im2double(imread('5.png'));
% gIm = SPDecolor(Im); % Default Parameters (sigma = 1e-2)
% figure, imshow(Im), figure, imshow(gIm);

Demonstration of SPDecolor

  • GcsDecolor: Gradient Correlation Similarity for Efficient Contrast Preserving Decolorization [Paper] [Code] [Slide]

  • Log-Euclidean Metrics for Contrast Preserving Decolorization [Paper] [Code]

  • Variable augmented neural network for decolorization and multi-exposure fusion [Paper] [Code] [Slide]

  • Color-to-gray Conversion on Wallpaper Dataset [Dataset]