Isaac Muse facelessuser

Interpolation with Kubelka-Munk Theory

Approach

This is an experiment showcasing interpolation via Kubelka-Munk theory. Approach generally employs the method as outlined in Spectral.js. Most of foundational topics are covered in Mixbox's paper, though Spectral.js uses an approach less focused on specific paints and instead opts to generalize and simplify the approach by simply generating reflectance curves directly from the spectral data. Results don't necessarily mimic specific paints, but give a pigment like feel.

	"""
	Calculate reflection curves for red, green, and blue that sum to the actual D65 white using SciPy.

	`calc_reflect_scipy.py` forces the curves to sum to a flat reflectance of 1. Integrating a flat
	reflectance under the sampled D65 illuminant does not land exactly on the color space's white
	(the spectral D65 and CMFs, at this sampling, are not quite `[0.3127, 0.3290]`), so that script
	rescales the rows of the XYZ to RGB conversion to compensate. That alters the color space.

	Here the color space is left alone. Instead of a flat 1, the curves sum to the smoothest white
	reflectance curve `w` that maps to exactly `[1, 1, 1]`, which is the color space's D65 white.

	"""
	Calculate reflection curves for red, green, and blue jointly using SciPy.

	This is a reference implementation of the joint solver in `calc_reflect.py`.
	The curves are the softmax of three logits (blue's logit is held at zero), so they
	always sum to exactly 1 and stay strictly between 0 and 1. The smoothness of the
	logits is minimized subject to red and green reproducing their target colors.
	Blue is implied: a flat reflectance of 1 is white, so blue is `white - red - green`.

	Requires `numpy`, `scipy`, and `plotly`.

	"""
	DLM color space.
	This is a prototype of the space as described here: https://prototypestudios.neocities.org/dlm
	We have not bothered to see exactly how they do their space, but have instead just implemented it as described.
	Any errors in implementation are likely our own.
	"""
	from __future__ import annotations
	import math
	from coloraide.spaces import Space, Prism, Luminant
	from coloraide.channels import Channel

	"""OkTCh color space."""
	from __future__ import annotations
	import math
	from coloraide import algebra as alg
	from coloraide import util
	from coloraide.cat import WHITES
	from coloraide.channels import Channel, FLG_ANGLE
	from coloraide.types import Vector
	from coloraide.spaces.lab import y_to_lstar, lstar_to_y
	from coloraide.spaces.lch import LCh

	"""
	Msh color space.

	Accounts for negative lightness and uses degrees for hue instead of radians.

	- https://www.kennethmoreland.com/color-maps/ColorMapsExpanded.pdf
	"""
	import math
	from coloraide import Color as Base
	from coloraide import stop

	## MIT License
	##
	## Copyright (c) 2023 Ronald van Wijnen
	##
	## Permission is hereby granted, free of charge, to any person obtaining a
	## copy of this software and associated documentation files (the "Software"),
	## to deal in the Software without restriction, including without limitation
	## the rights to use, copy, modify, merge, publish, distribute, sublicense,
	## and/or sell copies of the Software, and to permit persons to whom the
	## Software is furnished to do so, subject to the following conditions:

	"""

	Fancy lists in the style of Pandoc.

	---
	# A Python implementation of John Gruber's Markdown.

	# Started by Manfred Stienstra (http://www.dwerg.net/).
	# Maintained for a few years by Yuri Takhteyev (http://www.freewisdom.org).
	# Currently maintained by Waylan Limberg (https://github.com/waylan),

	# pragma: init
	from __future__ import annotations
	from coloraide.spaces.okhsl import Okhsl, okhsl_to_oklab, oklab_to_okhsl
	from coloraide.spaces.okhsv import Okhsv, okhsv_to_oklab, oklab_to_okhsv

	P3L_TO_LMS = [
	[0.4813798527499543, 0.4621183710113182, 0.05650177623872754],
	[0.2288319418112447, 0.6532168193835677, 0.11795123880518772],
	[0.08394575232299314, 0.22416527097756647, 0.6918889766994405]
	]