Skin Tone Pinning - Overexposed Skin Areas?

[mikemeister_admin]

Just to be different,HSB.The pins between 50 and75 were to hold the friend's face.Comparing them,I think Zog's RGB skin color is much better.

[mikemeister_admin]

By the numbers inRGB to get the eyes white,skinpin in lab then a little lizzard tail inHSB at quarter tone.friend pinned at halftone.We'll see if it makes any difference

[mikemeister_admin]

Ah ha - glad to see you have fallen, as I nearly always do, in concentrating on one aspect and blowing another!  The cabin wall colouring has really changed from the original and gone all blue-ish.

I think this says that for over-exposed smallish areas, we would really mask out the correctly exposed areas rather than try and do it with curves, especially things like eyes - what do you think?

PS I know we should be majoring (if that's the correct word) on curves, but surely not if it is not the best way of doing things.

[mikemeister_admin]

Full circle.RGB master and lowered lightness.I think the color shift took place with the skin pin ,Zog.

[billbrandi]

Thanks to all who responded.......

[-default]

> PS I know we should be majoring (if that's the correct word) on curves, but surely not if it is not the best way of doing things. <

I'm not sure that's possible.  No matter how good the results from a particular curve may be, there is always a better way to do it.  That includes better curves, including curves combined with masking, as well as techniques that do not involve curves.  That does not detract, IMHO, from using a particular image to show a set of techniques.

[mikemeister_admin]

Using CaptureNX,I put a control point (circle)on the face and used brightness and warmth checked against the hue clock.A little careful work with the healing brush would remove the hightlights on the cheek and nose.

[ggroess]

It still seems to be a bit off on the skin tones...unless you were going for the over exposure and not skin corrections.
The image on the left is the ACV file I posted..the Image in CM is yours from the site...

Many times more than 1 hue clock tells the story...

Greg

[mikemeister_admin]

Now I'm really confused - I thought GregM's skin was fine (as far as my eye was concerned) - so I copied it down to compare on my PC

I could not get any pink area to go less than 0 degrees - only the blow area.
Looking at the lower sample point (1st picture) 73,23,15 on mine and 72,23,16 for Greg's clock - I also tried different sample sizes but that didnt help.

Then I thought I could check out profiles, so I copied down Greg's image with the clocks on it and tried 3 profiles when the image loaded - sRGB, AdobeRGB & Leave it be - and monitored the clock colours, rather than the actual skin. 

I'm a little amazed at the results!

Note that the sRGB & As Is profiles give the same results - which means (?) that Greg had used sRGB when looking at GregM's image.

Why does nothing agree with Greg's clocks?

GregM, what happens when you try this experiment from your original image and the web posted version?

[mikemeister_admin]

PS. I've just downloaded the image I posted and compared it to what I had on my PC (just to see if posting would change things) and the images are identical - I used one on each layer with difference mode & then maximum levels adjustment.  The result gives a prefect black image.