One of the most noticeable differences between portraits taken outside using natural light as opposed to artificial light is the background. Images using artificial light tend to have darker backgrounds. This is crucial in catching the eye of the viewer and allows him or her to focus on the subject. This article is a guide in achieving this look using natural light only.
In most cases when shooting portraits with natural light only, the background comes out very bright. Technically speaking, when taking a photograph using natural light, the exposure is generally set for the subject’s skin tone, which is typically darker than the background. If achieving that dark background and having perfect lighting on the skin is important, the key is to underexpose the image. This will not only result in perfectly lit skin but it will also help retain the most detail. It is a lot easier to recover shadows than highlights. Contrary to popular belief, bringing out details from an underexposed RAW file does not mean creating noise when it’s done properly. In this day and age, any DSLR on the market can handle bringing out details from the shadows without creating noise. Using the following steps, it is guaranteed to walk away with phenomenal results.
Before I get to the steps, it is important to address a popular concern. Have in mind when looking closely at the final results; the subject continues to look underexposed. Parts of the skin were lightened to make it appear properly expose. This draws the viewer straight to the subject’s face, as it is the brightest part of the portrait.
Why Not Just Get the Exposure Right in Camera?
Technically, underexposing IS getting it right in camera. If the goal is to achieve that dark background that generally only comes with artificial lighting, it is less work in post-production to lighten the subject. There is no argument here; the skin makes up 10% of the image and the background is 90%. Lightening the skin is undoubtedly easier than darkening the background in post. Read on to learn why.
When shooting, it is important to shoot RAW. This file type contains all the extra information in the shadows. It’s possible to using Adobe Camera Raw or Lightroom, to brighten the shadows and make light pop. Personally, the main slider I use is the Shadows, Whites, Clarity and Luminosity ones under the HSL tab. Remember that underexposing the image and not properly lighting the subject are two completely different things. In order for this process to work, it is essential that the subject is properly lit. If the light hitting the skin is soft and evenly, brightening in post will be absolutely no trouble at all. When capturing the image, stay conscious of the light and the direction it is coming from. Once the subject is properly lit, underexpose the image SLIGHTLY.
Underexposing an image is no specific science. It varies from image to image, subject to subject and background to background. The best suggestion is to look at the images as they are taken and concentrate on the highlights. The goal is to avoid any blown highlights. If the highlights are blown, you can kiss the skin detail goodbye. Unless having overexposed Barbie skin is the objective, blown highlights should be avoided at all cost. In some cases, the background might have blown highlights like a sun flare or the sky. In instances such as those, overexposed pixels are fine.
An underexposed image makes dodging and burning three times easier. It’s a lot simpler to lighten the highlights on the skin than to darken the entire skin. When working with a darker complexion, shaping the skin and structure of the face is so much simpler too. Dodging and burning is vital if you want to have the perfect light in the final results.
I find the best setting for this method is to use Aperture Priority. I then override the camera’s auto settings by using the exposure compensation and underexposing with a few clicks. Every DSLR has this setting. I prefer to concentrate on the composition, light, pose and expression instead of wasting time and effort shooting in manual.
Using strobes is a fantastic method for lighting. I applaud all those who do it and do it well. However, I believe that when using strobes there are extra complications that can override a lot of the creative process. There is always a time and place for strobes. But when using natural light there is no need for an assistant, time is spent on composition, expression and communicating with the subject. As always, the methods used depend on a variety of factors. With this method of underexposing, using lights, reflectors and assistants is unnecessary to achieve amazing results. Natural light is quite powerful, one must take their time to master it and apply the correct methods for mind-blowing results.
Other before/after combos can be found on Dani Diamond Photography
Nice article. I will try this method on my next shoot.
Thanks for posting, will give this a shot
Great article, Dani! Do you mainly use Photo shop for post? Thanks!
Most likely, using lightroom is like playing basketball in a wheelchair.
I strongly disagree, they both have their uses. Lightroom excels at this type of work.
https://www.youtube.com/watch?v=0Vxjh6KJi8E
Nice article Dani and nice work too.
I would like to know that how you manage to get a catch light in eyes without strobe Or you add that in post?
I think when he recovers shadows the catch light in the eyes is more visible, after that, he only enhances it. I think this way because I've done it before.
Very nice article and I will try this. I don't think it will have me selling my lights however.
Exactly lol, its a nice tip for personal photography shoots but when shooting for commercial work it gets more complicated, Specially when you shoot continually for 8 hours, youll definitely need those flashes, which you just sold after reading this article.
Yup, Dani Diamond is my spirit animal.
Dani, You hit a home run with this article. I will be using this technique.
Amazing stuff as always Dani, I always love how well your portraits draw the eyes to the face
While I like the idea, I think this puts you at some risk as underexposing is a lot less forgiving than overexposing in terms of recovering details. And...I'd definitely rather have the control of lighting and not be recovering at all. Still looks excellent!
I usually experience the opposite; can pull far more detail from slightly underexposed images than overexposed images. Weird
Definitely weird, considering Digital sensors store way more information in the highlights. This is why "shooting to the right / expose to the right" is so popular. Detail is one thing, but there is always a cleaner result when correcting/adjusting highlights over shadows.
What I tend to experience (and I have read the petapixel article referred to a few posts below) is this: Whilst you can recover more detail from shadows they are generally noisier. You have less leeway/ room for error when recovering from highlights but the results are cleaner. I guess its a trade off as to what you prefer.
Yeah that *is* weird. Camera sensors are actually manufactured to store more data in the highlights (read here: http://www.luminous-landscape.com/tutorials/expose-right.shtml)
Perhaps you're not unlocking the full potential of your RAW converter to recover highlights, or you're overexposing TOO much. Try using your highlight clipping warning on your camera.
I disagree. Once details are clipped they're un-recoverable. The over vs. underexposed raw image test has been done many times and concludes that details are better recovered with under exposed images - http://petapixel.com/2009/06/09/underexposing-vs-overexposing/
Clipped is clipped in my book. And details in any clipped area won't come back. The fact is that there is more information in higher zones. So when recovering you will suffer less noise, banding, and color information if you are pulling a highlight vs. pushing a shadow.
Although a pulled image will look much better than a pushed image, I think he was referring to how it's easier to clip a highlight than a shadow.
3-stop pushed shadows will be noisy, muddy, and possibly de-saturated, but I can recover more stops of what information is there than I can in the highlights which clip white at around 1-1.5 stops ( with my Canon 6D at least)
It all depends on the camera I guess, my pentax k5 is awesome at pushing shadows at iso80 but the highlights can clip easy while my sigma Merrills shadows can get blotchy quick but it can store crazy amounts in the highlights (I tend to shoot the merrills at +0.7ev.)
Aha clipped is clipped so then I'll ask you something: in real life are there details in dark areas? Do you need details in the shadows or do you want to preserve as much detail in the highlights ? You and your stupid books. Learn it by doing and then read up on some techniques to refine it. If you increase the whites and shadows in post the picture will get a glow you would otherwise not have gotten. It took me a few days to realize that by trying out some things. Maybe you should do that as well.
Oh and I'll urge you to pull down overblown highlights to see for yourself what's going to happen. I'm coming from a landscape site and if you like a weird artificial sheen on the surfaces where the highlights are then go ahead and overexpose. recovering shadows is so much easier and it's logical that there is more detail than in highlights.
A test done on a 40D, 5 years ago. HARDLY significant today.
Robert, I agree with you. I perform this method and also take a few photos at different exposures that way I have myself covered.
Nice write up! thanks! You mentioned "the main slider I use is the Shadows, Whites, Clarity and Luminosity ones under the HSL tab" ...that gets you to a certain point then it's ready for the finishing touches correct? Are you dodging and burning? It almost looks like just dodging for the most part? That being said, do you expose to have the shadows in cam close to where you want to end up? Or...am I thinking about this wrong?
Could you make a video showing this? I learn better from seeing it in action. It would be nice to leave my lights behind for a day.
Please do. I came into this article looking for that, or some rundown of your post process, and instead the whole article could be boiled down to "1. underexpose your shots 2. bring up the highlights in post. 3. profit"
Yea the idea sounds great but it wold be wonderful to have a simple tutorial as an idea of how you approach it.
excelente, tratare de aplicarlo...
.
Nice, i've been doing this on some shots and i love it
It's articles like this that keep me coming back to fstoppers. Dani has some of the best advice I have come across and he inspires me to challenge new things. Dani Diamond my new fav photographer/retoucher/blogger!
Wonderful article Dani
I completely disagree. This is very bad advice. The LAST thing you want to do, even when shooting RAW is to deliberately underexpose in camera!
The *optimum* exposure for any RAW file is the one that puts the largest amount of light possible onto the sensor (and hence into the RAW file) without clipping the highlights. This is known as exposing to the right (ETTR) and will always give you a better and cleaner RAW file than exposing as the meter suggests, and certainly better than underexposing the whole scene as suggested here! If you absolutely *have* to underexpose to retain detail in the highlights in the background (eg. when shooting directly into a sunset) the of course, you have no other option (and this is why most of thee sunset shots are made with fill flash), but in most of the examples here there was absolutely NO NEED to underexpose in camera at all. The only thing that will be achieved by doing so will be to REDUCE shadow detail and INCREASE shadow noise!
The most sensible technique (other than fill flash or a reflector) to make the face 'pop' in images like these would be to expose to the right (ETTR) as much as possible to get the optimum exposure for the scene (and as much light on the subject as possible), then to drag the exposure of the entire image DOWN in Lightroom to the point where the background is suitably 'dark' (very much a matter of taste) and *then* to selectively pull back (lighten) the subject using the brush tool as mentioned in this article. That will ensure that the subject is properly lit to start with (not underexposed) and that the shadow areas on the subject (eg. dark areas beneath the chin, between the neck and long dark hair etc,) are not noisy and lacking in detail as they would be if the entire image was underexposed as suggested.here.
While this so-called 'advice' *may* seem to work for the OP, shooting with D800 (even though there's NO POINT whatsoever in doing it for all but one of the examples he's posted), for the VAST MAJORITY of DSLR owners it will simply mean they'll end up with underexposed, noisy files with no detail in the shadows. That's BAD ADVICE for 90% (at least) of the people reading this article and I really don't understand wy Fstoppers would publish something that will negatively impact the majority of people who might be inclined to give it a try.
ETA: Some articles about ETTR:-
http://en.wikipedia.org/wiki/Exposing_to_the_right
http://www.luminous-landscape.com/tutorials/optimizing_exposure.shtml
http://www.luminous-landscape.com/essays/the_optimum_digital_exposure.shtml
I was about to write something like that when I read your comment. Thanks for the clarification.
Are you a Canon shooter? I am and never agreed with this either... thats until Dani showed me one of his RAW files.
When I did i was like ohhhhh... ok i get it now. The dinamic range on the D800 allows for it and i agree for Dani's style shooting, slightly under makes sense.
It doesn't matter which camera you use, underexposing to throw away information in the shadows is just a bad idea.
And my point Dani is that regardless of the camera you're using, THROWING AWAY usable data by deliberately underexposing the RAW file simply to make the background *look* darker to start with is a ridiculous suggestion. All you need to do is ETTR as normal and then pull the whole exposure down in LR to get that darker look. This will preserve ALL the shadow detail you would have thrown away by underexposing in camera.
Why not just ETTR, pull the *whole* image down in LR (a 2 second operation) and then selectively 'lighten' the subject as you suggest anyway?
And yes, if you underexpose (even by half a stop) then you absolutely ARE throwing away data in the darkest shadows (unless maybe the DR of the whole image is shockingly small to start with).
I can see that you like to light 'flat' and then try to add contrast and mimic supplementary lighting by dodging some areas, but if you started out with a brighter file in the first place then your dodged areas would have more detail. How on earth could that be a bad thing? What's the downside? More data, more detail, less noise.... it's all good.
However, by contrast (no pun intended), the downside of your very bad 'advice' is that, for many people, following it will result in them having crappy, noisy, underexposed files with blocked up shadows and no usable detail in the darkest shadow areas. Not everybody shoots 'flat' with a D800 so even if this 'works' (as you allege) for you, it's not going to work for everybody and lots of people will end up messing up a lot of otherwise good images as a result.
This article is fundamentally flawed and it's a surprise to me that Fstoppers ever agreed to publish it.
If it works for him, and his work really suggests that it does, then what is the huge problem? Everyone shoots differently, and just because you don't think it's the best way does not make it the wrong way, nor does it make it unworthy of being published here. Using this technique certainly wouldn't stop a high end magazine from publishing the photo, or stop him from getting hired to shoot a brand campaign. It's the final image that counts, and Dani's level of quality is impeccable. Why on earth SHOULDN'T he share his technique with other photographer's who admire his work and are eager to become better photographers? The fact that you're willing to share another perspective on technique is respectable, but your entire attitude and stance on this subject seems very arrogant and selfish.
"Arrogant and selfish"? I would be more likely to call the OP's so-called advice 'arrogant and selfish', because while it *may* work for him with a D800 (even though there's no POINT whatsoever in doing it for all but one of the examples he's posted), for the VAST MAJORITY of DSLR owners it will simply mean they end up with underexposed, noisy files with no detail in the shadows. That's BAD ADVICE for 90% (at least) of the people reading this article, and that's why I'm opposed to it
.
"Arrogant and selfish"? I would be more likely to call the OP's so-called advice 'arrogant and selfish', because while it *may* work for him with a D800 (even though there's no POINT whatsoever in doing it for all but one of the examples he's posted), for the VAST MAJORITY of DSLR owners it will simply mean they end up with underexposed, noisy files with no detail in the shadows. That's BAD ADVICE for 90% (at least) of the people reading this article, and that's why I'm opposed to it
If it works for him and others, then I don't see why you have this crazy vendetta against it. Seriously, your rampage against this is ridiculous.