by Billy Biggs <vektor@dumbterm.net>
The following is a comparison of gradients generated in different colourspaces. The colours for the endpoints are defined in linear sRGB space. They were chosen such that none of the resulting colours were out of gamut in each of the colour spaces we tested with. They are listed in the table below. All of the gradients are generated assuming sRGB viewing conditions.
Source | Destination | |||||
---|---|---|---|---|---|---|
Red | Green | Blue | Red | Green | Blue | |
(0.0, | 0.0, | 0.0) | (1.0, | 1.0, | 1.0) | |
(0.1, | 1.0, | 0.1) | (1.0, | 0.1, | 0.1) | |
(1.0, | 0.1, | 0.1) | (0.1, | 0.1, | 1.0) | |
(0.1, | 0.1, | 1.0) | (0.1, | 1.0, | 0.1) | |
(0.1, | 1.0, | 1.0) | (1.0, | 0.1, | 0.1) | |
(1.0, | 1.0, | 0.1) | (0.1, | 0.1, | 1.0) |
Here are the linear in intensity gradients.
Here are the gradients not gamma corrected. These are the gradients you get when you interpolate the pixel values.
Ok I lied, I am just doing a cube and cube root, but by the CIELAB equations this is equivalent.
The following are gradients in the IPT colourspace by Ebner and Fairchild. See Development and Testing of a Color Space (IPT) with Improved Hue Uniformity from the IS&T/SID Sixth Color Imaging Conference in 1998.
Here are the gradients using my implementation of the CIECAM02 appearance model. See The CIECAM02 Color Appearance Model from the IS&T/SID Tenth Color Imaging Conference in 2002.