Skin tones

Tips, trick's, and links about computer graphics world

Moderator: joepal

Skin tones

Postby joepal » Sun Jan 02, 2011 6:29 pm

I've googled a bit for natural variations of skin tones, and found this formula for the RGB values for caucasian skin:

R = B * 1.5
G = B * 1.15
140 <= B <= 180

... which produces this table:


(I included the values above and below too).

Does anyone have similar formulas for asian and african skin tones?
Joel Palmius (LinkedIn)
MakeHuman Infrastructure Manager
Posts: 3699
Joined: Wed Jun 04, 2008 11:20 am

Re: Skin tones

Postby clausawits » Tue Jan 25, 2011 2:41 am

Interesting question... the link is not to a formula, and it's using CMYK, not RGB, but did you already see this: ... ction.html
They have a reference selection of tones at the bottom.

Again, CMYK, but from Howard Owen's post on
When I set up a scan for Caucasian people the Magenta trails the Yellow slightly and the Cyan in less than half of the Magenta.

If I sample a quarter tone area the CMYK might look like this: Cyan: 10 Magenta: 25 Yellow: 30 Black: 0

Asian people get a little more of everything plus a slightly higher separation between the Magenta and Yellow.

If I sample a quarter tone - mid tone area the CMYK might look like this: Cyan: 15 Magenta: 35 Yellow: 45 Black: 0

African Americans and people with darker skin tones get a slightly warmer treatment with enough cyan to keep it from going to red on press.

If I sample a mid tone area the CMYK might look like this: Cyan: 25 Magenta: 47 Yellow: 55 Black: 5

As usual, somebody always get's left out for example, people from India in this case...

I personally don't know enough about the whole thing, and I have never satisfactorily understood where "racial" boundaries lie... it might be easier to pick specific people (celebrities?) as reference points and just check several pictures of those people online.

For instance instead of "african", say Ving Rhames, or Venus/Serena Williams. Instead of "asian", Jet Li, or Zhang Ziyi. I may be pedantic, but dealing with explicitly defined sets is easier than poorly defined groups. For instance, both of my "african" examples were Americans. Both of my "asian" examples were Chinese. Maybe that's not what you think of. But saying specific individuals (of whom lots of photos are readily available for inspection/measurement/etc.) allows people to have a clearer common understanding of what's being discussed.

Posts: 45
Joined: Sun May 18, 2008 10:36 am

Return to Off topic

Who is online

Users browsing this forum: No registered users and 1 guest