How to visualize colors?

[achetverikov]

I want to visualize color interpolation similarly to how it is shown in the docs here

https://facelessuser.github.io/coloraide/interpolation/

How to do this?

[facelessuser]

As ColorAide is mainly meant for practical applications of colors, I don't ship such tools with the library. They are more documentation and development, with that said, the tools are all available in the repo. While the library is rigorously tested, the tools are not, just an FYI. So, they can break on occasion without me knowing.

The tool that generates these is called tools/interp_diagram.py. It's built on top of the tools/slice_diagram.py tool. Basically, you give it a list of colors and define the slice. If the colors do not fall on the slice plane, they will be adjusted such that they do. You need the tool requirements installed: https://github.com/facelessuser/coloraide/blob/main/requirements/tools.txt.

Here we specify 3 Oklab colors and specify that we want to use the monotone interpolation method. We set a title as well. Then we
specify the X and Y axis slice range which is the a and b component respectively, and we give them a range of -0.4 - 0.4. We set the constant with k to the l component with a value of 0.7. The gamut is assumed to be the srgb gamut, but you can change it it desired.

python3 tools/interp_diagram.py -s oklab -c "oklab(0.7 0.15 0.1)" -c "oklab(0.7 -0.05 0.1)" -c "oklab(0.7 -0.09 0.02)" -c "oklab(0.7 -0.03 -0.12)" -x "a:-0.4:0.4" -y "b:-0.4:0.4" -k "l:0.7" -T "Monotone Interpolation in the Oklab Color Space" -t "Lightness at 70%" -m 'monotone'

This can be a bit restrictive, so we do have 3D interpolation diagrams as well. As long as the color space is restricted to 3 color components, you can show them . This one is a bit easier to use as you don't have to specify slice bounds, just the color space and the colors and the output (display) space to view it in.

python3 tools/interp_3d_diagram.py -c 'blue' -c 'yellow' -c 'orange' -c 'purple' -s srgb -d srgb -m monotone

[facelessuser]

I should also mention that I don't make promises that I won't revise how any of the tools work at a moment's notice. They are mainly there to facilitate whatever I'm currently developing or wanting to visualize, so if I end up wanting to visualize things in a different way, I may adjust them. I do occasionally refactor, fix and adjust to better handle things.

[achetverikov]

Thanks, this is extremely helpful. I ended up using the 3d version and got this nice plot:

[facelessuser]

Very cool, you can see it has some cleanup still needed "displayed in 'None" space", but it works pretty well 🙃.