Vibrance is the concept of increasing saturation/chroma without affecting the high saturation area. It is better than saturation for the reason that it avoids overblowing areas that are already saturated. This allows users for quick manipulation of saturation within image without hurting the higher saturation area. The easiest method to increase vibrance of a image is to use a inverted saturation mask, and only change the saturation level and luminosity is to be preserved.
There is actually some explanation over this thread: Help: How to achieve "Vibrance"? - G'MIC - discuss.pixls.us
What do artists think about having a Vibrance filter as opposed to HSx filter? Could the HSx filter be extended to use a mask to avoid affecting high saturation area?
I have studied it before。
Description of Photoshop:
Vibrance: Adjusts the saturation so that clipping is minimized as colors approach full saturation. This setting changes the saturation of all lower-saturated colors with less effect on the higher-saturated colors. Vibrance also prevents skin tones from becoming oversaturated.
This is the description in the photodemon Code:
'Photoshop pioneered the concept of a “vibrance” adjustment in CS4. Vibrance is similar in concept
’ to saturation (as you can probably infer from the name), but unlike saturation, it changes color
’ “vibrance” in non-linear ways. Already-vibrant colors are largely ignored by the tool, while largely
’ unsaturated colors are also ignored. Tones with middling saturation receive the largest changes,
’ which is what allows the tool to produce more “realistic” output compared to linear saturation
'The algorithm PhotoDemon uses has undergone a number of revisions. At present, it automates an
’ S-curve adjustment to the underlying image’s saturation (via the HSL space, specifically - not HSV).
’ This provides reasonably good control, while limiting the amount of change applied at the high and
’ low ends of the scale.
I now realize it by adjusting the HSL curve of GMIC：
There are two gifs. The file is too large, so I upload it here:
It looks pretty useful.
For anyone wanting an already usable workflow, I could replicate the effect using Krita’s Cross-channel adjustment curves:
However note that since it’s based on HSL and not HSY’, it doesn’t actually keep the luminance - even in the example shown, the mouth area is much darker in the “High Vibrance” version than in the “Low Vibrance” one (and the Original is somewhere in the middle).
The settings are pretty simple, just note that both Channel and Driver Channel are “Saturation”:
This is of course Low Vibrance option.
EDIT: To be more truthful to PS, you’d need to adjust the curve to be more like a hill or a hole in the ground instead of straight line. That way it will only adjust the values in the middle of Saturation range.
I think it is the HSV model.