build-color 1.0.2

Python color-science workbench for HDR tone mapping, appearance models, Delta E, spectral, ICC, and LUT workflows

Build Color

Python color-science workbench for perceptual spaces, HDR tone mapping, color appearance models, Delta E metrics, chromatic adaptation, spectral utilities, and ICC profile generation.

Project Telos | gather | crucible | index | forum | telos | emet | buildlang

Build Color is a Python color-science workbench for people who need more than hex conversion. It covers perceptual spaces, HDR tone mapping, color appearance models, Delta E metrics, chromatic adaptation, spectral utilities, ICC profile generation, and an optional GUI.

Quick Start

pip install ".[all]"
build-color

Launch the GUI, or use the CLI:

build-color info ff6030
build-color convert ff6030 --to oklab
build-color difference ff0000 00ff00 --metric all
build-color harmony ff6030 --scheme triadic
build-color spectrum --temp 6500
build-color icc --gamma 2.2 --output display.icc

Features

Color Spaces (15+)

sRGB, Linear RGB, XYZ, xyY, CIELAB, LCH, Oklab, Oklch, JzAzBz, JzCzhz, ICtCp, HSV, Display P3, BT.2020, Adobe RGB, ACEScg

Tone Mapping (12 operators)

ACES (Narkowicz + Hill), AgX (neutral/punchy/golden), Reinhard (simple + extended), Hable/Uncharted 2, Lottes, Uchimura/Gran Turismo, PBR Neutral (Khronos glTF), BT.2390 EETF, BT.2446 Method A, custom knee function

HDR Processing

PQ (ST.2084) encode/decode, HLG (BT.2100) encode/decode, BT.2390 EETF tone mapping, BT.2446 HDR-to-SDR

Color Appearance Models

CIECAM02 forward/inverse (machine-epsilon roundtrip), CAM16, CAM16-UCS uniform color space, hue quadrature

Color Difference (7 metrics)

CIE76, CIE94 (graphics/textiles), CIEDE2000, CMC(l:c), JzAzBz Delta E, Oklab Delta E, HyAB

Chromatic Adaptation (9 methods)

Bradford, CAT16, CAT02, Sharp, Von Kries, CMCCAT2000, Fairchild, Bianco-Schettini, XYZ Scaling

Additional

• Spectral rendering - Planck blackbody, CIE 1931 CMFs, daylight illuminants, SPD-to-XYZ integration
• ICC profiles - Generate ICC v4 display profiles from primaries + gamma
• CVD simulation - Protanopia, deuteranopia, tritanopia, achromatopsia (Brettel et al.)
• Gamut mapping - Clip, soft compression, Oklab chroma reduction
• Color harmony - Complementary, triadic, tetradic, analogous, split complementary, monochromatic

GUI

Interactive workbench with 5 tools:

• Color Inspector - Pick a color, see it in every space with luminance, wavelength, contrast
• Palette Generator - Generate harmonious palettes, copy as CSS
• Color Difference - Compare two colors across all metrics with visual grades
• Tone Mapping - Preview all 12 operators with curve visualization
• CVD Simulator - See how colors appear to color-blind viewers

build-color gui

CLI Commands

Command	Description
build-color	Launch GUI (default)
build-color info <color>	Show color in all spaces + metadata
build-color convert <color> --to <space>	Convert between color spaces
build-color difference <c1> <c2>	Compute color difference
build-color harmony <color> --scheme <type>	Generate palette
build-color adapt <xyz> --from D65 --to D50	Chromatic adaptation
build-color spectrum --temp <K>	Blackbody spectral data
build-color icc --gamma 2.2	Create ICC profile
build-color gui	Launch GUI workbench

Installation

pip install ".[all]"    # Everything (numpy + scipy + PyQt6)
pip install .           # Core only (numpy)

Python API

import numpy as np
from build_color.spaces import srgb_to_oklab, oklab_to_srgb
from build_color.tonemap import aces_filmic, pq_eotf
from build_color.difference import delta_e_2000
from build_color.adaptation import adapt, ILLUMINANTS
from build_color.appearance import ciecam02_forward, ViewingConditions

# Color space conversion
oklab = srgb_to_oklab(np.array([0.8, 0.2, 0.1]))

# Tone mapping
hdr = np.array([0.0, 0.18, 1.0, 4.0, 10.0])
sdr = aces_filmic(hdr)

# Color difference
lab1 = np.array([50.0, 25.0, -10.0])
lab2 = np.array([60.0, 20.0, -5.0])
de = delta_e_2000(lab1, lab2)

# Chromatic adaptation (adapt XYZ from D65 to D50)
d65 = ILLUMINANTS["D65"]
d50 = ILLUMINANTS["D50"]
xyz = np.array([0.4596, 0.2985, 0.0614])
adapted = adapt(xyz, d65, d50, method="cat16")

# CIECAM02
vc = ViewingConditions(white_point=d65)
appearance = ciecam02_forward(xyz, vc)
print(f"J={appearance.J:.1f}, C={appearance.C:.1f}, h={appearance.h:.0f}")

Architecture

build_color/
  spaces.py       15+ color space conversions
  tonemap.py      12 tone mapping operators + PQ/HLG
  appearance.py   CIECAM02, CAM16, CAM16-UCS
  difference.py   7 color difference metrics
  adaptation.py   9 chromatic adaptation methods
  spectral.py     Planck, CIE CMFs, daylight
  icc.py          ICC v4 profile generation
  blindness.py    CVD simulation (4 types)
  gamut.py        Gamut mapping
  harmony.py      6 color harmony schemes
  cli.py          Command-line interface (8 commands)
  gui/            PyQt6 interactive workbench

See USAGE.md for an install/build line, the full command and API reference, and worked examples with expected output.

License

Build Color is released under the FSL-1.1-MIT. The source is source-available: you may read, run, modify, and build on it for any purpose other than a competing commercial use, which is reserved to the Licensor to fund continued development. Copyright (c) 2022-2026 Zain Dana Harper. All rights reserved. See LICENSE.