derekfountain,
it looks nice, but om my screen to dark.
It is ealy to make it lighter and giving it more contrast.
I look against a screen with 113 candela/m2 a normal screen lightening.
So is your screen much lighter , what you see, you produce ?
Frits
A skin tones challenge
-
- Posts: 251
- Joined: Fri Jan 26, 2007 1:24 pm
it looks nice, but om my screen to dark.
It is ealy to make it lighter and giving it more contrast.
I look against a screen with 113 candela/m2 a normal screen lightening.
So is your screen much lighter , what you see, you produce ?
You're right, it is a bit dark. Since the photo was taken in the shade, I thought the darkness with the bright sunlight in the background looked OK. But this morning I played with the Lab lightness curve a bit to raise the shadows, and it does look better. It can't stand that much, however, because there's quite a brightness range in the skin tones of the face. Boosting the curve to give a bit of contrast always seems to make somewhere important go flat.
I also wiped around the hat brim with the sponge in desaturate mode to remove that green halo. I could do more, but I have to move on with my life!
-
- Posts: 4927
- Joined: Fri Sep 20, 2013 8:29 pm
You're right, it is a bit dark. Since the photo was taken in the shade, I thought the darkness with the bright sunlight in the background looked OK.
Always with backligthing take a stop wider, or take a spot reading.
I could do more, but I have to move on with my life!
Who not ? Sometimes I spend to much time on a photo to make it right.
But the goal of that is making the photo right, but the big goal is mastering it.
To get it in the fingertips how to do this or that.
When you now all the tricks ( I am still learning) it is a peace of cake.
Frits
What makes this image so interesting, and difficult, is the mixed lighting. This is evident on the boy's shirt, which shows magenta light coming from the left, and yellow light from the right. In such cases, I generally turn to RGB first to see whether I can remove the competing casts without resorting to a mask. Not this time. In this case, I found that I needed a mask, but surprisingly it was to split the top of the image from the bottom, not left and right.
It took two passes; an RGB curve to get the bottom half of the image, and an Lab adjustment for the top half, combined with a simple layer mask to blend the two together.
There is more room for improvement, certainly, but this is not bad. With few exceptions, all of the areas of the face are between 12 and 1 on the hue clock. The right side of the boy's face is still at one o'clock, more yellow than I would like.
I've included the curve files I used. Apply the RGB one first, duplicate to a new layer, and then apply the Lab curve.
It took two passes; an RGB curve to get the bottom half of the image, and an Lab adjustment for the top half, combined with a simple layer mask to blend the two together.
There is more room for improvement, certainly, but this is not bad. With few exceptions, all of the areas of the face are between 12 and 1 on the hue clock. The right side of the boy's face is still at one o'clock, more yellow than I would like.
I've included the curve files I used. Apply the RGB one first, duplicate to a new layer, and then apply the Lab curve.
-
- Posts: 4927
- Joined: Fri Sep 20, 2013 8:29 pm
-
- Posts: 4927
- Joined: Fri Sep 20, 2013 8:29 pm
Yes,
that is my problem. Must I add more mangenta or do always the first thing the usual Highligt, neutral and blackpoint.
The problem is that I have lots of scanned dia-positives withe nearly the same problems.
Besides the problem with dark dia-positives witch I think could solve so Greg mentioned, I have such problems as in this picture.
So, I have seen the boy lean backwards. so I correct this first.
After that I do the usual (H),N,B.
By doing that, there is automaticly added mangenta to the picture.
Strange thing is, that there is natural neutral area on the left of the hat.
When I set 2 neutrals, one on that particular point and one on the white stripes of the shirt, it will not work in RGB and get I a strange effect.
So the face is now not yellow, but to mangenta.
Frits
( delayed by my grandchild)
that is my problem. Must I add more mangenta or do always the first thing the usual Highligt, neutral and blackpoint.
The problem is that I have lots of scanned dia-positives withe nearly the same problems.
Besides the problem with dark dia-positives witch I think could solve so Greg mentioned, I have such problems as in this picture.
So, I have seen the boy lean backwards. so I correct this first.
After that I do the usual (H),N,B.
By doing that, there is automaticly added mangenta to the picture.
Strange thing is, that there is natural neutral area on the left of the hat.
When I set 2 neutrals, one on that particular point and one on the white stripes of the shirt, it will not work in RGB and get I a strange effect.
So the face is now not yellow, but to mangenta.
Frits
( delayed by my grandchild)
-
- Posts: 4927
- Joined: Fri Sep 20, 2013 8:29 pm
-
- Posts: 4927
- Joined: Fri Sep 20, 2013 8:29 pm
Return to “Want us to work on one of your images?”
Who is online
Users browsing this forum: No registered users and 9 guests