Opaline ·  writing  ·  color science

the light
on your face.

how she reads a face in daylight — undertone, value, contrast — and turns it into a palette that belongs on you.

methodology · vol. I · spring MMXXVI

color analysis is older than the app — it's a craft from the ateliers, refined by colorists, now quietly run by a model that sees a thousand faces a week. here's how she does it.

i. one selfie, daylight.

opaline asks for a single photo, face to a window, no filter, no makeup if you can help it. that's the whole capture. she doesn't need a calibrated card or a color checker — modern phones and a decent window do the job.

ii. three measurements.

she measures three things, in this order: undertone (warm, neutral, cool — read from the lab a*/b* axis of the skin sample, not the surface tone), value (how light or dark the face sits on a perceptual lightness scale), and contrast (the distance between hair, skin, and eye in value).

undertone is read in CIELAB, not HSV — LAB is perceptual, which means the numbers behave the way your eye does. value uses a perceptual lightness curve calibrated against the Monk skin tone scale, not Fitzpatrick — Monk has ten steps instead of six and reads more accurately across deeper tones.

iii. a season, not a stereotype.

the three measurements drop into a deterministic season tree — the same logic a colorist uses, written down. you come out soft autumn or bright winter or one of the twelve, not because the model decided, but because the numbers say so. the season is the door; what we do with it is the room.

iv. what she does with it.

v. where to read more.

we'll publish the full methodology paper at launch — including the season tree, the LAB thresholds, and a side-by-side against three other consumer tools. until then, the short version above is the entirety of what she actually does.