More neutral than almost all other gray cards

[Asher Kelman]

Selecting skin colors (for a mask) is relatively straight forward with a selective color selection, and the colors can be changed at will with a Color Balancing filter. That would also allow to Color Correct the camera's response to a mix of different skin colors in the same image. It can even be used in an action if speed of operation is more important than absolute perfection.

With this method we can also apply a partial correction, or if we want to add a tan or sunburn to an existing image, which is often considered to be more pleasing (to a customer) than accurate skin colors.

Bart

Bart,

This is obviously a great approach and one that Picto color uses. You are suggesting using the color selection tool. How would you set up a batch process. Would you do one skin type at a time and can one store such selections as one would have to specify tolerance sufficient to distinguish a caucasian from an Asian or Black skin, each with sub types.

Asher

 

[Andrew Rodney]

From the ColorRight Max site:

"The COLORRIGHT MAX gray patches are guaranteed to be more neutral than almost all other gray cards on the market."

Bottom line. The CC is a good WB reference (lightest gray) and a very handy internal reference but do not be fooled into thinking that it will match the numbers that you see published. Even the numbers that Gretag (now X-Rite) publish are acknowledged to be averaged.

My first question, coming into this late and skimming the posts is, what are we referring to in terms of neutral? The target itself? Easy to measure with a Spectrophotometer. The "results" of the capture?

The most neutral and highest L* target I've measured is the Bable Color target.

The values the CC provide from X-Rite (GMB) is the measured values from a Spectrophotometer. Its not what you'll find from resulting pixels (although you can shoot for those output referred pixels using various techniques). That isn't necessarily anything like the scene referred colorimetry although some say it should be.

In the end, having an output referred CC that corresponds with the measured values gives us colorimetrically correct images of CC charts, and not necessarily anything else we'd photograph.

 

[Doug Kerr]

Hi, Bart,

Interesting essay. Thanks.

Here is a (horribly lit) scene which could be considered to have a correct general White Balance:

Indeed the lighting must have been quite "interesting". On this "corrected" image, we see the "neutral" patch on the McB CC showing very nearly a reference white chromaticity, while what we assume to be a BabelColor White Balance Target at the lower left shows something rather removed from that.

Best regards,

Doug

 

[Doug Kerr]

Hi, Andrew,

My first question, coming into this late and skimming the posts is, what are we referring to in terms of neutral? The target itself? Easy to measure with a Spectrophotometer.

And we know Drew has a reflective spectrophotometer - he used it to measure the transmissive spectral performance of the original Color Parrot white balance diffuser.

Of course there's always the matter of what measurement geometry is most pertinent to this application.

Best regards,

Doug

 

[Bart_van_der_Wolf]

This is obviously a great approach and one that Picto color uses. You are suggesting using the color selection tool. How would you set up a batch process. Would you do one skin type at a time and can one store such selections as one would have to specify tolerance sufficient to distinguish a caucasian from an Asian or Black skin, each with sub types.

A selected color range can be saved, and one can add multiple skin colors in a single selection (the dropper tool with a plus, in the Select Color Range dialog). One could save such a range with a generic name, e.g. "SessionSkinColors". That would allow to use a generic script which loads that specific color range for the session at hand. The selection will act as a mask when the action creates a Color Balance layer with some highlight color adjustments for the specific camera.

Bart

 

[Bart_van_der_Wolf]

Hi, Bart,

Interesting essay. Thanks.



Indeed the lighting must have been quite "interesting". On this "corrected" image, we see the "neutral" patch on the McB CC showing very nearly a reference white chromaticity, while what we assume to be a BabelColor White Balance Target at the lower left shows something rather removed from that.

Yep, but such is life, compromises are inevitable.

In fact, the reason that the BabelColor target does't agree with the WhiBal and Graycard and mini ColorChecker in this particular setup, is the angle of its surface (reflecting some cloudless blue sky up ahead) to the lightsource (direct sunshine through tree branches). I was short a hand to angle it properly.

Actually, the BabelColor target would be more accurate for WB-ing the surface of the skin on the back of my hand, but I chose to balance on the less blue direct sunshine instead. The skin is not going to be accurate anyway, due to IR pollution.

Bart

 

[Asher Kelman]

Nothing is pure! Neither grey cards nor threads. I've copied Bart's excellent opening to software options for controlling and getting desirable color to the thread for Software Discussion. We can have in that way a rich coherent resource that will help find the options we have for color management for portraits and events. We want white balance and pleasing skin color and the latter for individuals of different skin type.

So add comments here as you wish but put your richest ideas on software in the new dedicated thread.

The evaluation of new tools is so important. Drew's invention is intriguing to say the least. I too want ease of use. I am looking forward to my copy of the device so I can check it out. If it works in some situations then that would be a good thing. It certainly looks like a nifty device.

Asher

 

[Doug Kerr]

Hi, Bart,

Yep, but such is life, compromises are inevitable.

In fact, the reason that the BabelColor target does't agree with the WhiBal and Graycard and mini ColorChecker in this particular setup, is the angle of its surface (reflecting some cloudless blue sky up ahead) to the lightsource (direct sunshine through tree branches). I was short a hand to angle it properly.

Indeed.

This reminds us of the need, with "mixed" lighting (in the "tidy" case where the subject is planar), to have the "receptor" of our incident light "pickup" system (the neutral target, or the face of a measurement diffuser) parallel to the subject surface (notwithstanding all the instructions to the contrary found in the Farmers' and Photographers' Almanac).

And if the subject is not planar (and, especially in glamor photography, we usually hope for it not to be), and we have a significant "mixed light" situation, there is no white balance color correction that will completely deal with the incident light chromaticity issue over the entire subject, and there is no such thing as the "best" orientation of the receptor (unless it is, "parallel to the surface of that part of the subject we wish to have 'best corrected' ".

Best regards,

Doug

 

[Bart_van_der_Wolf]

This reminds us of the need, with "mixed" lighting (in the "tidy" case where the subject is planar), to have the "receptor" of our incident light "pickup" system (the neutral target, or the face of a measurement diffuser) parallel to the subject surface (notwithstanding all the instructions to the contrary found in the Farmers' and Photographers' Almanac).

Absolutely, in my book as well, anyway. The frontal surfaces, as seen from the viewer, are the most important for our overall color impression. From that the rest follows the ambient environment in a natural way.

And if the subject is not planar (and, especially in glamor photography, we usually hope for it not to be), and we have a significant "mixed light" situation, there is no white balance color correction that will completely deal with the incident light chromaticity issue over the entire subject, and there is no such thing as the "best" orientation of the receptor (unless it is, "parallel to the surface of that part of the subject we wish to have 'best corrected' ".

Indeed, the non-planar non-parallel surfaces will take on whatever ambient cast is available, and that's completely natural.

Bart

 

[Ben Rubinstein]

Problem with Bart's solution is that it takes a lot of time, hardly appropriate for post processing of a wedding with hundreds of proofs.

However the underlying problem shouldn't be there should it? If the WB is correct then both the skin colours and general hue/tint should be accurate. If it is not it's because the profile is incorrect I assume unless parts of the image are lit with a different temperature light than others, a problem not really relevant to this discussion as there can be no global solution to such a problem other than control of the lighting.

 

[Deleted member 55]

The skin is not going to be accurate anyway, due to IR pollution.

Hey Bart!

Is IR/UV pollution in your opinion a posable cause for "muddy skin color" when shooting a street scene?

I have a friend that blames "muddy skin color" on a brand of strobe when he can not use white balance to get rid of it. I think he might have an older non UV coated flash tube and this adds to the problem when shooting outside!

What is your opinion on this? Could this be the case?

 

[Michael Tapes]

Bingo! A WB offset will change everything.

That's how it should be done. One should WB for the dominant illuminant, then CB for a (local) range of colors. A single click 'reference' can't [always] do both at the same time.

Bart

I concur with my addition in brackets.

Michael Tapes

 

[Drew Strickland]

That's it, Michael?

No more defense of your upcoming G7 product?

 

[Doug Kerr]

That's it, Michael?

No more defense of your upcoming G7 product?

Mee-ow!

Best regards,

Doug

 

[Doug Kerr]

Hey, Drew,

No more defense of your upcoming G7 product?

I heard the G7 will have a special patch for fixing up Eskimo skin tone. Could open up a whole new market.

Best regards,

Doug

 

[Michael Tapes]

Mee-ow!

Best regards,

Doug

Darn.....Just had my nails cut! Bad timing....





...and in the neutral corner, wearing the red trunks.....
Miiiiiiiiiichaaaaaaaaaeeeeeeeeeeeel TAAAAAAaaaaaaaaaaaaaaappppppppppppppes

 

[Drew Strickland]

I heard the G7 will have a special patch for fixing up Eskimo skin tone. Could open up a whole new market.

Best regards,

Doug

Yeah, could be. Except considering we have a pending patent on the whole "skin tone selector using wb" concept I doubt he would want to go there.

 

[Michael Tapes]

This keeps getting better and better.

 

[Drew Strickland]

This keeps getting better and better.

Alright, since Michael isn't going to do it.

Here goes yet another round of attempting to get the point of this tool across.

Since my training is in an Applied Science field, where things like practicality get factored into the equation, I thought I would put together a little equation for the seemingly large number of "pure" science folks on here.

Without further ado the two equations battling for supremacy in this thread...

BS = (25)E + (25)S + (100)Q

MS = (90)E + (90)S + (98)Q

 

[Drew Strickland]

By the way. Please don't tell my wife I am "wasting" more time on this thread. She'll kill me. She wants me to do practical and important things like figuring out how to market these tools and running an actual business. Something about keeping food in the pantry. Shhh...

 

[Bart_van_der_Wolf]

Hey Bart!

Is IR/UV pollution in your opinion a posable cause for "muddy skin color" when shooting a street scene?

I have a friend that blames "muddy skin color" on a brand of strobe when he can not use white balance to get rid of it. I think he might have an older non UV coated flash tube and this adds to the problem when shooting outside!

What is your opinion on this? Could this be the case?

Hi Will,

I don't think it is UV. The Silicon sensitivity at the short end of the visual spectrum is limited to something like 350-380nm or something like that, and lenses also filter UV light. Besides, it would be easy to check with a UV or even Skylight filter if there is any contamination coming from there.

If there is a suspect, then IR could be it. It of course also depends on the camera, how well is it IR filtered, if at all? B+W have a filter that can be placed on the lens and will reject (UV and) IR. It's their number 486 filter. However it has 2 drawbacks; 1. it's expensive, and 2. it introduces a lens cast at wider angles. It's price will be prohibitive if one just wants to try if it makes a difference.

The first thing to do however, is to make a better default calibration in Adobe RGB if that's the editor of choice. The default profiles for e.g. Canon cameras are notoriously bad for anything red(reddish). A Macbeth ColorChecker and a free script (I use e.g. Rags Gardner's script, and Simon Tindemans script also has a good reputation). One can even tweak based on that better baseline to take particular care of the skintone patch, but if IR is the culprit then it's a moving target practice under variable light.

Bart

 

[Drew Strickland]

Alright, since Michael isn't going to do it.

Here goes yet another round of attempting to get the point of this tool across.

Since my training is in an Applied Science field, where things like practicality get factored into the equation, I thought I would put together a little equation for the seemingly large number of "pure" science folks on here.

Without further ado the two equations battling for supremacy in this thread...

BS = (25)E + (25)S + (100)Q

MS = (90)E + (90)S + (98)Q

Oh yeah, you probably need to know what the variables are. Sorry. Here:

BS= Bart's Solution
MS= MAX Solution

E= Ease of Use
S= Speed
Q= Final Image Quality as judged by Pleasantness to Average Client

 

[Drew Strickland]

So, as you can easily see. For the BS solution to win one would have to assign a value of essentially 0 to both ease of use and speed.

Of course, your weightings of each variable may differ. But, I hope this helps to illustrate the point a little more clearly.

 

[Doug Kerr]

Hi, Drew,

Alright, since Michael isn't going to do it.

Here goes yet another round of attempting to get the point of this tool across.

Great.

Since my training is in an Applied Science field, where things like practicality get factored into the equation, I thought I would put together a little equation for the seemingly large number of "pure" science folks on here.

Dunno for sure who those all are. My formal training is in engineering, which as you surely know, was for many years called "applied science". But that seems to be something different from what you are advocating.

Without further ado the two equations battling for supremacy in this thread...

BS = (25)E + (25)S + (100)Q

MS = (90)E + (90)S + (98)Q

Well, I'm not sure why you found it necessary to use quotes around the coefficients. Must be some kind of mathematical notation only understood by you Applied Science (as evidently distinguished from engineering) types.

So, where's the part that explains the point of the tool?

Best regards,

Doug

 

[Doug Kerr]

Hi, Drew,

Oh yeah, you probably need to know what the variables are. Sorry. Here:

BS= Bart's Solution
MS= MAX Solution

E= Ease of Use
S= Speed
Q= Final Image Quality as judged by Pleasantness to Average Client

Well, I would think that for these two "weighting" equations to make sense, they should be normalized (that is, all the coefficients should sum to 1, or since for some reason you seem to be working on a percent-like scale, to 100.)

Best regards,

Doug

 

[Doug Kerr]

Hi, Drew,

Yeah, could be. Except considering we have a pending patent on the whole "skin tone selector using wb" concept I doubt he would want to go there.

Ah, yes, very nice. (Kerr makes "toward me" motion with his hand, as if encouraging an opposing witness, telling more than he means to, to keep going.)

Best regards,

Doug

 

[Doug Kerr]

Hi, Michael,

This keeps getting better and better.

Thank God, it will all be over tomorrow night.

No, wait - that's another baloney train. Sorry I got confused.

Best regards,

Doug

 

[Drew Strickland]

Well, I'm not sure why you found it necessary to use quotes around the coefficients.

I thought it would be fun. That's how they taught it when I was in grade school way back in the 1930's.

So, where's the part that explains the point of the tool?

Best regards,

Doug

You mean other than the massive time savings and ease of use?

Um...

I thought I would dance around that a little longer.

 

[Drew Strickland]

No, wait - that's another baloney train. Sorry I got confused.

Best regards,

Doug

So what name does your baloney have?

McSame or Hussein?

 

[Doug Kerr]

Hi, Drew,

I thought it would be fun. That's how they taught it when I was in grade school way back in the 1930's.

You mean other than the massive time savings and ease of use?

Compared to doing what?

If I want to make the whole image a little cooler, I just push the color temperature slider over a little, whereas with the ColorRight MAX - no wait, I can't use that to make the whole image a little cooler.

Sorry, I lost my head.

Well, most people are likely to want their skin a little redder anyway.

Best regards,

Doug