That really is your call. In my view you are improving a shot that had some significant issues with it and it seems to be improving with each iteration.
I know photographers that will work on an individual image in PP for hours if not days to get it right while others who will toss any image that isn't nearly perfect straight out of the camera. If this is the best shot from your trek and you really want to pull out what you can, keep on going. If you don't really care, then by all means go onto something else.
I have some shots that I took some years ago that I did not want to throw out, but my PP skills were not good enough to get them up to a quality level that I found acceptable. That group is getting smaller every year, as my skills have improved and the PP tools have gotten better.
I go with what Manfred is saying, but ............. and it's one of the hardest skills to learn in photography ............. knowing when to say "No, it's not good enough" and binning it, is one of the key development stages for any photographer who really wants to push themselves in terms of quality.
The other side to that same coin (and I do it all the time) is to look back at those things I thought were superb 4 - 5 years ago and think that they wouldn't reach muster now given the standards I set for myself.
If you think that Flash Fill would have been a good idea, then, I think that a better approach for your Post Production would be to emulate Flash as Fill as a first stage in that Post Production process.
*
Below here is an A/B comparison of your first Image and your second redo of it. Note, whilst ‘skin tone’ has been addressed (a second time) in the second redo, the girls face is still in shadow and the shot appears backlit with no fill. This (lack of) lighting renders the girl’s face depthless and does not enhance her character.
*
Imagine if you did have two off camera flash units each with a soft bounce modifier – you could have filled both sides of her face to accentuate the soft cheek bones and lighten the eyes and the teeth and add some richness to the lips and hair and create a sense of depth to the Portrait.
This is a ROUGH and QUICK example
Below here are seven stages: 01 to 03 expands the RANGE of the shadow area in small incremental stages; 04 selectively dodges and burns the face, hair, eyes, lips and teeth to create the illusion of two fill lights; 05 is a colour correction (by eye and to taste using a studio monitor, no reference skin tone); 06 enhances the Mid-Tone contrast (always lacking in back-lit available light portraiture); 07 final sharpening and resizing.
*
Below here is an A/B comparison of the original image and the ‘Final’ redo:
Having achieved the "Flash Fill" Post Production, you are then free to correlate skin tones to other shots (for example of the same girl in various lighting scenarios) and/or tweak the skin tones to you personal artistic liking.
WW
Yes, I agree.
And yes, I agree.
*
An additional comment to Post #24:
Post 24 was an example of where (and why) to BEGIN the PP process to get the outcomes that the OP seems to want. It certainly was not meant as an example of how to rescue an image. The image doesn’t appear to have been set up, with any particular Post Production in mind.
*
This shooting scenario was described as “it was a trek and the shots are reactions to a moment.”
Part of the learning process, especially for candid portraiture is to:
1. Be prepared to "react to the moment"
2. Shoot with the intent of PP
3. Shoot safe.
If the emphasis is on learning, and it seems that it is - then even before we get to PP, we need to seek out if we were best prepared to "react to the moment".
In this regard we’d look at seeking to find the exact reasons for choosing: Spot Metering; the use of Av Mode; setting ISO125 (which rendered Tv = 1/100s for this shot).
*
When reacting to THIS particular moment, we’d probably decide in an instant that the most relevant area is the young girl’s face – so then we must ask – what did we spot meter and how did we adjust to get the safest shot possible for Post Production, because we knew that we did not have a flash to fill the backlit Portrait.
*
That’s easy to write. But not as easy to do.
But I can absolutely positively guarantee that that it can never be done without trying to do it.
And the first step is always ask yourself, especially as the lighting scenarios change: "is my camera best prepared to react to the moment?"
WW
+1 to what Bill has written.
When shooting in these "opportunistic" situations, having the camera on hand and ready to shoot (with the right lens and other equipment), set up with the appropriate settings (including focusing and metering modes) is critical for success. His other piece of advice of shooting for PP is equally important, as these types of shots virtually always need a bit of tweaking in post.
I find not cropping too tightly when shooting helps as well as it gives you enough space around the edges to do a bit of straightening and cropping. I try to get multiple shots, changing my camera settings a bit along the way when I do these opportunistic shots. Note - I don't use burst mode, although some people do, I just like a bit more control when I do shoot.
Just as an aside, I do a lot of this type of shooting when I am traveling. You do have to be aware of your surroundings and understand the location and the lighting to be successful. Above all, practice lots as with any other skill, practice is what helps you get those killer shots.
Thank you all for taking the time (lots of it in some cases) to help me get better. It is indeed a long process and I knowthat all of your efforts are not wasted. I have picked up a lot just from this thread although only practice at how to do much of whet you suggest will make my results improve significantly.
The toughest part is looking at a photo and deciding what it needs and that only comes from experience and listening to advice. Again, thank you.
Personally Stephen, the first thing I think I would have done is address the white balance on this shot. There is a pretty good cast happening here and once dealt with would make follow-up in post a heck of a lot easier.
Not arguing with that as a first step, but asking for the procedure to do it: specifically what would be the reference to address a White Balance?
I see a + Magenta cast dominant on the shirt; a + Yellow cast dominate on the face; and an overall minor + Green cast.
Also, the kicker is, we do not know the (base) colour of the shirt, we can assume white, but it might not be so; in which case although the + Magenta cast in the shirt might be correct, there might be also a minor + Cyan cast, if for example the base colour of the shirt is pink.
I could be wrong, but I think there seems to be a lot of reflected (colour) light bouncing into the room and then onto various areas of the Subject: I don't think the casts we see are are result of a simple WB 'error' which can be easily corrected by removing one of them.
WW
thanks all for an interesting discussion!
This should help get you started (first one I came to) but there is a lot of info out there regarding white balance referencing. Some video on YouTube as well but I didn’t want to burn my data plan watching them.
Good eye. However, I think the shirt is less a magenta reflected light issue and more a laundry issue, but I wasn’t there.
You could be but I also don’t think there is a simple 'error' that can be easily corrected myself. In fact to bring this shot to where I would want it if it were mine would involve a fair bit of work with the color aspects. But I think the shot would be worth the effort.
Last edited by Loose Canon; 19th April 2016 at 10:50 AM.
Bill - how can there be both a green and magenta cast in the same image? The two are complementary, so from a technical side, I can see one or the other but not both. If this were an old partially faded colour print, perhaps, but in a digital image? What am I missing in my colour theory?
I definitely see the yellow in the face and I also pick up a bit of green. I suspect that the magenta is actually close to the true colour of the top she is wearing.
andThis should help get you started (first one I came to) but there is a lot of info out there regarding white balance referencing.
Indicated by these two responses the meaning of post #29 has been misunderstood.how can there be both a green and magenta cast in the same image? The two are complementary, so from a technical side, I can see one or the other but not both.
I haven't the time at the moment to: clarify my meaning; better articulate the points that I was making; and reshape the question that I was asking, but I will do so shortly.
Sorry for any inconvenience pertaining to the lack of clarity in my Post #29.
WW
If it helps, here is the original, straight out of the camera, no post processing. She is standing in the shadow of a structure with no walls but with a roof. The day is reasonably warm and bright and she is getting as cool as she can. The bamboo pole is one of four, each one holding up a corner of the roof.
Yes it sure does help, so does the description of the shooting location and I will expand my later response because of you posting this - thank you very much.
But is that the JPEG File SOOC? if so then it is not without post processing because the JPEG is processed according to the "Picture Styles" set in the camera - if it is a JPEG do you know what the "Picture Styles" were set to? (pp90 to 95 EOS 60D User Manual). This is only a minor point so don't go to any great effort.
Already noted AWB (Auto White Balance) was set.
WW
OK Ta
Thank you for the link. As mentioned my question was not clear and I was not seeking instructional material as to how to make a White Balance before beginning Post Production, sorry for the lack of clarity.
What I meant was: What would one (you) use inside that scene as a reference for the "White Balance".
My point being, and as I mentioned, there appears to be quite a few ‘casts’ in different areas of the image, probably caused by reflected light.
And in regard to the apparent conditions of the image in discussion, I disagree with the opening sentence of the commentary in the link that you provided, viz (my bold and underlined for emphasis):
The point is that statement will hold true and everything will be all fine and dandy when setting a White Balance if the scene has a predominately a CONSISTENT Colour Temperature – AND – the scene is reasonably free from direct and indirect coloured lighting affectations.Setting your white balance is one of the most important edits you make in Camera Raw because if you do it right, your color will be spot on, and you won’t have any color correction problems to deal with later in Photoshop.
So, if we agree (and it seems we do agree) that the scene has a complexity of colour 'casts' and probably those casts are a result of reflected light from coloured objects, mostly outside the scene affecting the main subject.
Additionally, and now almost certainly confirmed by the production of the JPEG image SOOC, we were referencing an image already having gone through Post Production . . .
When I asked what would we set as a reference for a WHITE BALANCE . . . I was attempting to make the point we would not be aware as to what degree and affect reflected colour light and also any PP would have on the ‘casts’ that we were seeing - so I was asking "what would we consider safe to use as a reference for our White Balance?"
***
Yes, I see your point.
My response above should explain much of what I meant.
But additionally, to address your specific question, I was using the word ‘cast’ in a liberal sense to include any and all (induced) colour aberrations that might be possible – and (importantly) these seem to occur in different parts of the image.
As you and Terry point out, the shirt is probably rendered as close to the correct colour and there is no really strong ‘cast’ as such, on it.
What I was meaning was: ‘there appears there could be a magenta cast on the shirt, but how do we know what colour it actually is?’
In other words there could be a colour irregularity in the shirt, there probably is a colour irregularity in the face (+yellow) there seems a general colour irregularity in the background lit area and the edges of the bamboo (+green) . . . remember these comments are all in reference to the already Post Produced images were were viewing.
Moreover, I think that there has been confusion caused by lack of definition of two words:
Firstly, I used the word ‘casts’ in a liberal meaning – NOT as a Colour Temperature / Myriad Shift “CAST” across the image as a whole.
Secondly, I interpreted “White Balance” when it was used in Terry’s commentary, as a Technical Term meaning "White Balance" (as it relates to setting a White Balance for a scene or an image file with even or mostly even Colour Temperature.
I note that in Terry’s response, he mentions and agrees with my analysis concerning the complexity of the multiple colour issues in the scene and he notes that he would take the time and effort to correct them – this I would refer to as “Colour Correction” or “Colour Balance” and these have a different meaning and different procedure to “White Balance”.
***
Thanks again for posting those details and the images SOOC.
Below here is an A/B of the SOOC file and the first image you posted in this thread. Note that the Post Production of the image has introduced (won’t use the word ‘cast’) colour affectations the face and the bamboo (+yellow) and on the shirt (+ magenta). So we now have ascertained that some (most?) of the ‘casts’ we have been discussing in this thread - are a result of Post Production.
Assuming that:
1. the AWB has done a reasonable job; (and that’s often quite a safe assumption if it has some daylight and a reasonable amount of natural surrounds, i.e. all the background in your scene – but you can test that statement out by referencing you own experiences over several hundred shots)
2. in the knowledge that Canon’s ‘Picture Style’ set to “Standard” does not make radical adjustments apropos White Balance (I’ll ask you to take my word on that, but, if you have a passion you can test all the Picture Style combinations and make a grid of outcomes for different lighting scenarios. That is quite time consuming and really not all that necessary unless you have a passion for trivia and experimentation – AND/OR – you want, or need, to capture and then use the JPEG file SOOC - OR - if you are setting up the JPEG FILE because you want to use the HISTOGRAM and BLINKIES as CRITICAL TOOLS for review and analysis whilst still shooting in situ)
3. Then: I conclude that we can be reasonably safe in using the “White Balance” of the JPEG SOOC as our (your) reference and you could proceed to nuance the “Colour Balance” of individual elements of the scene to suit your Artistic Taste and any Technical Compliance as you see fit, but overall nuance the Colour Balance to enhance your VISION of the Final Image.
Thanks for waiting for this more detailed reply – I hope this is presented more accurately and is of more benefit to you.
WW
When I post processed this in PS, I used among other things some adjustment layers for levels, curves, contrast and vibrance. Maybe the vibrance boosted the colour and introduced the casts.
If I may ask a question from an ignorant button presser.
How do you recognise a cast? For me the only way to recognise a cast is to compare the photo to the original scene, which even I (who was there) cannot do as my memory does not keep such intricate details. So what is everyone doing, are we comparing with what our head guestimates is the "right" colour? Even looking at the SOOC shot does not help as the camera may not have saved the colours correctly, right? And if we are comparing with some imagined ideal, then everybody will have a different opinion on what is the correct adjustment.
Another question is that if people are asking for advice from the experienced photographers out there, should the non processed shot be posted as well as the processed one to make the critiques easier (and perhaps more correct).
Steve
Last edited by scully; 21st April 2016 at 06:59 AM.
Yes. Likely.
Levels and Curves could also, if done individually for R:G:B.
Answering for myself only -
When viewing the first image that you posted and my viewing it on a standardized studio (photographic editing) monitor in room with standardized low level room/studio lighting - I saw:
> an unusual yellow in the bamboo and the cheeks of the face;
> an unusual RANGE OF DIFFERING degrees of magenta in the shirt;
> a green tinge across the background scene.
These nuances of colour I recognized as a result of looking at many images over several years and quite a few years working in a professional photographic darkroom and more recently also having been taught and mentored by a Digital Post Production expert.
***
Yes. Probably the JPEG SOOC is not precisely correctly - I mentioned that – but in most cases as I also mentioned AWB is very close for the general Lighting Scenarios that I described.
Also (and this is likely the MOST most important point) it is a very rare circumstance where one ever requires absolute or close to absolute Colour Accuracy – for some Insurance and Forensic purposes are the only two which I have encountered over 40 or so years.
In these situations there is necessity for rigorous control over lighting; colour temperature; emulsion; development; (or digital sensor calibration) and digital post production (and/or neg printing and then the final image MUST be viewed in a standardized environment.
For all other purposes “exact and accurate” Colour Balance is not necessary, what is often necessary however is CONTINUITY of the Colour Balance throughout a series of shots of the same subject and this is often quite difficult to attain/maintain if the Subject moves between different lighting scenarios.
A common example of this difficulty to maintain the CONTINUITY of Colour Balance across different Lighting Scenarios is found during Wedding Portraiture: for example when the Bride’s Maids wear GREEN or PURPLE gowns – this is so for both Film and Digital capture.
***
Yes, each of us will have a different interpretation of what “correct” is to be.
This is especially so and especially evident if we all are viewing the same image on computer screens, many of which are not standardized and all are in different room lighting.
***
Yes. I have often asked for exactly that or commented about that. Most recently a few days ago at Post #10 here: “Could you load the raw file to a drop box?”
WW