This photos isn’t bad, but looks too pale…
2022-06-26–12.07.26.CR3.zip (21.5 MB)
2022-06-26–12.07.26.CR3.xmp (25.2 KB)
This file is licensed Creative Commons, By-Attribution, Share-Alike.
2 Likes
2022-06-26–12.07.26.CR3.xmp (9.1 KB)
1 Like
fine!
Maybe a little to yellow/grenish. But I did a very fast edit.
Edit: This might be somewhat better:
2022-06-26–12.07.26.CR3.xmp (9.5 KB)
4 Likes
Nice challenge!
I did not crop the vignetted upper left corner to leave the church hall intact and not to cut the foot of the kneeling gentleman
2022-06-26–12.07.26_RT-3.jpg.out.pp3 (15.0 KB)
1 Like
you may use colorbalancergb to add vibrance, contrast and linear color grading or perceptual saturation
There are two aspects of this scene that, in my opinion, make the photo pale:
-
The illumination is very diffuse and even. Diffuse light is actually very good for portraits because you don’t have harsh shadows, but if it’s too even (without soft shadows), it lacks the necessary depth.
-
The color of the light and the walls (with corresponding diffuse reflections) is very similar to the color of the skin, which also makes it difficult to separate people from the background.
The task here is to make this separation afterwards.
By white balancing I first got the color separation between people and background. Then I lightened the skin color a bit. Contrasts increased and with the help of channel mixer and color balance rgb tried to imitate the original color mood of the room.
2022-06-26–12.07.26_02.CR3.xmp (13,5 KB)
darktable 3.9.0~git1798.baa74228-1
7 Likes
I decided to try a different approach to give this photo more punch. RAW extracted using Filmulator (+1EV). In GIMP I used my saturation_h_m_l.py plug-in (as mentioned in the thread on vibrance) to boost the areas of medium saturation - particularly flesh-tones.
1 Like
Working on a new warm-ish RT preset (I know, I know…) at the moment and I kind of like how that one came out. Thanks! 
RT5.8 dev
2022-06-26–12.07.26.jpg.out.pp3 (19.5 KB)
You are far more analytical… I cheated… I took a guess the floor might be grey and include any color cast. WB on an average of the floor…the dog looked pretty good and the skin tones seemed okay so I went with it
In my job as an event photographer, I struggle with such lighting conditions very often:
With hundreds of such photos you slowly develop the view and the rutine in handling. 
2 Likes
Nice scene and tricky to handle the colors.
Here’s my try in RT_dev:
2022-06-26–12.07.26.CR3-1.jpg.out.pp3 (14,8 KB)
2 Likes
Ya nothing beats experience…here I might try to cheat again on the desktops just to see where it lands or work to a pleasing skin tone…I often find just fixing or getting a good skin tone yields a nice overall wb for the whole image… but again I am a rank amateur…
2022-06-26–12.07.26.CR3.xmp (16.4 KB)
Alternative version without shadows/highlights, this one is just white balance as shot, rgb per channel tone mapping, unsharp mask and LAB vibrance
2022-06-26–12.07.26_b.CR3.xmp (13.2 KB)
“In my job as an event photographer, I struggle with such lighting conditions very often:”
From my professional photography experience I would suggest you can save a lot of grief by getting a photographic grey card or a color checker card and taking a photograph with that under the lighting conditions you are working under. Takes but a moment to do. Then this would let you workout the white balance correctly for the development of the whole shoot which you could save as a preset or style in DT.
2022-06-26–12.07.26.CR3.xmp (10.6 KB)
Here is my version using DT version 4.10 which is soon to be released
- Lens correction
- Increase exposure to +1.0 EV
- Set Filmic using auto tune levels eyedropper
- Used the add basic colorfullness preset in the color balance RGB module
- In the options tab of filmic I experimented with preserve chrominance option and decided on luminance Y
- In the look tab of filmic I increased the contrast slider a little
- I also increased the extreme luminance slider to about 25%
- Experimented with white balance and manually set to 2800K with a tint of 0.895
- Applied local contrast at default values
- Used the preset sharpen demosaic (AA filter) for the diffuse or sharpen module
- Used the tone equalizer to do so exposure tweaks
1 Like
