| Colour | I | P | T | CIPT | hIPT° | MHunt |
|---|---|---|---|---|---|---|
| Primary | — | — | — | — | — | — |
| Comparison | — | — | — | — | — | — |
| Colour | J | C | h° | M | s | Q |
|---|---|---|---|---|---|---|
| Primary | — | — | — | — | — | — |
| Comparison | — | — | — | — | — | — |
| Colour | J′ | a′ | b′ | C′UCS | h′° | ΔEUCS |
|---|---|---|---|---|---|---|
| Primary | — | — | — | — | — | — |
| Comparison | — | — | — | — | — |
| Colour | L* | a* | b* | C* | h° |
|---|---|---|---|---|---|
| Primary | — | — | — | — | — |
| Comparison | — | — | — | — | — |
Compare ΔEIPT, ΔEUCS, CIEDE2000, and ΔE₇₆ across average, dim, and dark surround conditions for the current colour pair.
| Surround | ΔEIPT | ΔEUCS | CIEDE2000 | ΔE₇₆ | IA | IB | JA | JB |
|---|
CAT16 vs CAT02 — Chromatic Adaptation Transforms
Both CAT02 (CIE 159:2004) and CAT16 (Li et al. 2017, CIE 248:2022) are von Kries diagonal chromatic adaptation transforms. They differ only in the 3×3 sharpening matrix. CAT16 was derived to reduce gamut-compression distortions observed with CAT02 under certain illuminant pairs (especially tungsten→daylight). It also avoids the negative cone-sharpening excursions that CAT02 can produce for highly saturated spectrally narrow stimuli.
In practice, most sRGB colours shift by less than 1 ΔE between CAT02 and CAT16 under D65 — but the difference becomes visible for non-D65 illuminants at high D.
CAM16 Forward Model (Li et al. 2017)
CAM16 is a direct replacement for CIECAM02 using CAT16 instead of CAT02. The pipeline: XYZ → CAT16 partial D scaling → MCAT16−1 → HPE LMS → sigmoidal compression → opponent channels (a, b) → achromatic response A → J, Q, C, M, s, h. Prediction accuracy is ~0.14 ΔE00 better than CIECAM02 across RIT-DuPont, Witt, and Leeds datasets.
Hunt Effect — Colourfulness & the FL Factor
The Hunt effect (Hunt 1952, 1994) describes the psychophysical observation that colours appear more colourful at higher luminance levels, even when chromaticity is identical. FL = 0.2k&sup4;(5LA) + 0.1(1−k&sup4;)²(5LA)⅓ where k=1/(5LA+1). Colourfulness: M = C · FL0.25. At LA=1 cd/m², FL≈0.02; at LA=1000, FL≈0.42.
CAM16-UCS J′a′b′ Uniform Colour Space
CAM16-UCS maps CAM16 correlates into a perceptually uniform space using lightness compression (cJ=0.007) and colourfulness compression (cM=0.0228). ΔEUCS = √(ΔJ′²+Δa′²+Δb′²). Euclidean (invertible) — a key advantage for gamut mapping and rendering pipelines. Error residuals comparable to CIEDE2000.
iCAM Framework Evolution: iCAM → iCAM06 → iCAM08
iCAM (2002): Bradford CAT, IPT, first spatial image appearance
model.
iCAM06 (2007): CAT02, IPT, refined tone reproduction,
ΔEIPT, Hunt/Stevens effects.
iCAM08 (2008+): CAT16, IPT + CAM16-UCS, improved spatial
adaptation, colourfulness restoration step.
At point-scale (this tool), the spatial step collapses to global CAT16 + CAM16 correlates — demonstrating per-pixel behaviour at patch level.
ΔE Comparison: ΔEIPT vs ΔEUCS vs CIEDE2000
ΔEIPT:
√(ΔI²+ΔP²+ΔT²) — simple Euclidean,
excellent hue uniformity, invertible.
ΔEUCS:
√(ΔJ′²+Δa′²+Δb′²) —
CAM16-based, perceptually uniform, invertible.
CIEDE2000: weighted, 5 correction terms, best psychophysical fit,
not invertible.
ΔE₇₆: CIELAB Euclidean, simplest.
Applications: HDR, Gamut Mapping, Camera ISP, Material You
- HDR tone mapping — spatially adaptive white maps preserve local contrast.
- Gamut mapping — constant-hue rays in PT/J′a′b′ avoid hue rotation.
- Cross-media matching — predict shifts from print to display to projection.
- Camera ISP — model observer adaptation for white-balance and colour correction.
- Material You (HCT) — Google draws from IPT hue-uniformity principles.
CAT16 Chromatic Adaptation Transform (Li et al. 2017)
[-0.250268, 1.204414, 0.045854],
[-0.002079, 0.048952, 0.953127]]
Degree of adaptation D:
D = F · [1 − (1/3.6) · e−(LA+42)/92] ∈ [0, 1]
Adapted cone responses:
Rc = (D·Yw/Rw + 1−D) · R
Gc = (D·Yw/Gw + 1−D) · G
Bc = (D·Yw/Bw + 1−D) · B
CAM16 Forward Model
R′a = 400 · (FL·Rc/100)0.42 / [(FL·Rc/100)0.42 + 27.13] + 0.1
Opponent channels:
a = R′a − 12·G′a/11 + B′a/11
b = (R′a + G′a − 2·B′a) / 9
Correlates:
J = 100 · (A/Aw)c·z
C = t0.9 · (J/100)0.5 · (1.64 − 0.29n)0.73
M = C · FL0.25
s = 100 · (M/Q)0.5
Q = (4/c) · (J/100)0.5 · (Aw+4) · FL0.25
CAM16-UCS Uniform Colour Space
M′ = ln(1 + cM·M) / cM [cM=0.0228]
a′ = M′·cos(h) b′ = M′·sin(h)
ΔEUCS = √(ΔJ′² + Δa′² + Δb′²)
Hunt Effect — FL Luminance Adaptation Factor
FL = 0.2·k&sup4;·(5·LA) + 0.1·(1−k&sup4;)²·(5·LA)⅓
M = C · FL0.25
IPT Colour Space (Ebner & Fairchild 1998)
MHPE = [[ 0.38971, 0.68898, -0.07868],
[-0.22981, 1.18340, 0.04641],
[ 0.00000, 0.00000, 1.00000]]
2. Power-law compression (γ=0.43):
L′ = sign(L)·|L|0.43
3. IPT rotation:
MIPT = [[ 0.4000, 0.4000, 0.2000],
[ 4.4550, -4.8510, 0.3960],
[ 0.8056, 0.3572, -1.1628]]
ΔE Colour-Difference Formulas
ΔEUCS: √(ΔJ′²+Δa′²+Δb′²)
CIEDE2000: weighted L′C′H′ with 5 correction terms
ΔE₇₆: √(ΔL*²+Δa*²+Δb*²)
Known Limitations
- Point mode only: Spatial iCAM08 pipeline (image input, Gaussian blur local whites) not implemented.
- sRGB gamut boundary: 8-bit per channel. No wide-gamut or HDR input.
- Single-state adaptation: No temporal adaptation dynamics.
- No fluorescence model.
- Browser floating-point: IEEE 754 double precision. <10−10 from reference.
CAM16 & CAT16
[2] CIE 248:2022. “The CIE 2017 Colour Fidelity Index for accurate scientific use: Rf.” Includes CAT16 specification.
iCAM Framework
[4] Fairchild, M.D. & Johnson, G.M. (2002). “Meet iCAM: A next-generation color appearance model.” IS&T/SID CIC10, 33–38.
[5] Ebner, F. & Fairchild, M.D. (1998). “Development and testing of a color space (IPT) with improved hue uniformity.” IS&T/SID CIC6, 8–13.
CIECAM02 & Colorimetry
[7] CIE 15:2004. “Colorimetry.” 3rd edition.
[8] Sharma, G., Wu, W., Dalal, E.N. (2005). “The CIEDE2000 color-difference formula.” Color Research & Application, 30(1), 21–30.
[9] IEC 61966-2-1:1999. “Default RGB colour space — sRGB.”
Hunt Effect & Colour Appearance
[11] Luo, M.R., Cui, G. & Li, C. (2006). “Uniform colour spaces based on CIECAM02 colour appearance model.” Color Research & Application, 31(4), 320–330.
About this Tool
Enter multiple hex colours (comma or newline separated). The engine computes pairwise ΔEIPT, ΔEUCS, CIEDE2000, and ΔE₇₆ for every combination using current illuminant, viewing conditions, and D. Minimum 2 colours.
The P×T canvas superimposes both CAT02 (open circles) and CAT16 (filled dots) positions for each colour. The dashed connector shows the adaptation shift between the two matrices. For D65 sRGB colours, the shift is typically <1 ΔEIPT. For non-D65 illuminants at high D, the shift becomes significant — use this to quantify the practical difference between iCAM06 and iCAM08 adaptation.
The FL vs LA canvas shows how the Hunt luminance adaptation factor varies over the full luminance range (0.01–10000 cd/m²). Colourfulness M=C·FL0.25 is scaled accordingly. The vertical marker tracks the current LA setting. Use this to study how colours lose colourfulness in dim environments (cinema) and gain colourfulness outdoors.
Future: implement the full spatially-varying iCAM08 pipeline accepting image input. The image would be Gaussian-blurred to estimate local adaptation whites, enabling per-pixel CAT16 adaptation before the IPT/CAM16 transform. This enables research into local adaptation, HDR tone mapping quality, and colourfulness restoration after tone compression.