An example of a shot using the pseudo-HDR technique. The same picture in higher resolution can be look at me at 500px.
HDR (High Dynamic Range) stands for high dynamic range. Typically, photographers refer to 'HDR' as 'HDR images' or 'HDR photos' taken using exposure brace technology. Fork (fork exposure or bracketing exposure) means obtaining the same image with different exposures and then combining them into one image that has a high dynamic range. Many examples of pictures taken with 'classic HDR technology' can be found here.
But there is a fairly simple way to get HDR-like result from just one RAW shot... The raw data storage is often sufficient for significant shadow and highlight correction. The essence of the pseudo-HDR method is to 'develop' a RAW image with different corrections exposure, and then collect the developed images into one, which will have an extended dynamic range.
I sometimes use this pseudo-HDR technique:
- I make 11 JPEG files with correction from one RAW file exposure -5 -4 -3 -2 -1 0 +1 +2 +3 +4 +5. If the dynamic range of the used camera is very small, for example, Nikon D40 it is in the region of 9 steps, then I use a set of corrections exposure -2 -1.66 - 1.33 -1 -0.66 -0.33 0 +0.33 +0.66 + 1 +1.33 +1.66 +2. For each individual camera, you can choose your optimal set of shifts exposure... More or fewer shifts can be used, but it is desirable to have a symmetrical set with plus and minus corrections.
- I stitch the resulting JPEG files into one pseudo-HDR image. Most often I use the rather simple Google Nik Collection HDR EFEX PRO 2 plugin. Now the Nik Collection plugin package is paid (it was bought from Google by DXO), but at one time it was distributed for free... The plugin works with Adobe Photoshop and Adobe Lightroom. You can also stitch files using other methods, programs and plugins. Who does this - describe your methods and programs in the comments.
Example
First, I take the original RAW file and convert it 'as is':
Original shot with + -0
Then I do bracketing exposure 1/2 step, five minus steps and five plus steps. I save each step to a JPEG file and end up with 11 shots.
Bracketing exposure in 1/2 stop increments (11 shots)
Next, I load the images into Nik Collection HDR EFEX PRO 2 by calling the appropriate menu from the automation in Adobe Photoshop:
Loading JPEG files
Then I use Nik Collection HDR EFEX PRO 2 tips and achieve the result I need:
Plugin interface
As a result, I get the result I need:
The result of the work. The horizon line has been deliberately centered to emphasize the symmetrical reflection of clouds in the water
This method can sometimes be very well applied not only for landscapes, architectural photography, but also for portraiture, in which it is not so easy to execute bracketing exposure without moving the model in the frame, examples:
On modern cameras with a very wide dynamic range, such as the Nikon D850, sony a7r iv etc. this method can give very good results. But with modern cameras, and the usual correction of shadows, midtones and highlights results in decent pseudo-HDR without the need to do exposure bracketing.
Comments on this post do not require registration. Anyone can leave a comment. Many different photographic equipment can be found on AliExpress.
Material prepared Arkady Shapoval. Training/Consultations | Youtube | Facebook | Instagram | Twitter | Telegram
IMHO it is practically pointless at today's stage of development of editors, they today allow you to do the same by simply pulling the sliders - the same DxO has a tutorial on this https://www.dxo.com/project/creating-an-hdr-effect-from-a-single-raw-image/ ... Plus, Sony has an in-chamber D-Range Optimizer, which works exactly the same as I understand it. But if you have the opportunity to compare two approaches (from the same file or creating "bracketing") - it would be interesting to take a look.
Theoretically, by correcting the ranges in Raw, you can, of course, achieve an even better effect, especially if you also correct the ranges themselves for the task or do it in the curves. But the method described above is remarkable first of all because it can be automated by a script and give on average the same good results like painstaking work with sliders and ranges.
probably it was worth putting out at least one raw file so that you could play with it and compare it with your results
No problem, here
aha, thanks a lot!
there is one more nuance that confuses.
sensors with a large DD, usually do not like overexposure and have a large margin of shadow pull. so a scene with really big DD on them is usually shot with a large exposure compensation in minus to preserve the light.
accordingly, generating a symmetrically corrected jpeg from such a file does not seem to be practical advice.
These are the nuances that I mentioned. For example, the S5 pro's lights stretch perfectly. The method can be customized for a specific camera.
> usually do not like overexposure and have a large margin of shadow pull
Depends on the camera, apparently: for example, the ETTR (expose to the right) rule helps me quite well on (relatively old) Fuji and Pentax carcasses: the sky is in place, and the shadows without noise in the case of landscapes.
Sony also works well at 20-24 MP. And vice versa - on ancient Olympus, if you have to choose, I'd rather underexpose.
I think it is quite realistic to find out the limit for your camera, when information in the shadows / lights is irretrievably knocked out.
PS although - in the case of some lenses - under- or overexposure = frame in the trash. True, then bracketing saves :-)
twisted the example - it is completely incomprehensible why to take so much.
you just bring up the shadows in the raw and you get exactly the same “HDR”.
and there are a lot of lights in the frame, which does not fit with the idea of getting HDR)
Look, it's not just that I indicated “Pseudo-HDR” in the title of the note.
It's written
But with modern cameras, and the usual correction of shadows, midtones and highlights results in decent pseudo-HDR without the need to do exposure bracketing.
hdr with variable aperture - you can see that the shadows outside the depth of field are in the “haze”, although the ripples on the water and the reflection are clear. HDR + 0- 1 stop in constant exposure and ISO mode, variable aperture. Thanks to the variable depth of field (0 on hyperfocal), it was possible to achieve natural image reproduction (just glued in HDR).
Interesting application
And a really interesting trick. Thanks for sharing
Why is this necessary? hdr variable aperture?
What mode do you shoot hdr in?
Here is my example hdr without changing the aperture.
Well, the exposure is determined by two parameters - aperture and shutter speed. ISO, formally, does not affect exposure, the amount of light remains the same, the gain changes. So, for exposure, it is usually customary to play with shutter speed, providing under-light and overexposure for it. But nothing prevents you from doing aperture bracketing for exactly the same purposes. An interesting idea, you will have to try and see what happens. As a result, an underexposed picture will be sharper due to a greater depth of field, an overexposed picture, on the contrary.
Why is this necessary if there is HDR on the camera?
Which one?
Well, I have an asus 3 years ago. he has )
Since when did Asus release cameras? o_O
ever since cameras got into smartphones
I also have it on Olympus e-m10, but I don't use it. The pentaxes also seemed to have K-5 and younger
Canon Rp has HDR for example, but it is only for architecture, tk. takes three frames.
During this time, the model has time to move.
Canon D700 ma HDR mode (3 frames).
700d
Asus does not produce cameras. Laptops, smartphones, PCs and monitors, watches and various stuffing for them. The commentator means, as I understand it, the ZenFone 6 smartphone. It has a 48mm retractable camera and the ability to shoot HDR. It is sad when people do not distinguish the camera from the camera in the smartphone ((q
Dear experts, do I understand correctly that this method does not work with just the native Nikon Capture NX-D converter?
Works. Also, you make an exposure from raves with export to jpeg. Then sew with something
This is the main point of the question - it is impossible to stitch with Nikon Capture NX-D, as far as I understand. Can you tell me the available version of the "stapler" to test the method?
To be honest, I have not seen a normal HDR stapler)) I am usually not satisfied with the result, maybe my hands are not from there. There is Photomatix Pro - it works pretty well, both as a plugin for Photoshop, and separately.
HDR as a rip-eyed "processing" went out of fashion already when I first became interested in it. The familiar effect of instantaneous tonal range compression is very difficult to achieve. And it is not necessary, in principle.
In general, processing really comes down to adequate manipulation of only three sliders during development - exposure, blacks, whites. If necessary, contrast in minus on images with hard light or plus on low-contrast ones, followed by fine-tuning of blacks, whites, exposure. If still not enough, you can try shadows / highlights, but this time as a tool of last resort - it gives you that disgusting pseudo-HDR effect. The problem is that blacks / whites manipulate the headroom for shadows and highlights that is present on a 12-14-bit equal when trying to display them on an 8-bit monitor, while shadows / highlights work with an 8-bit remainder, with masks. leaving this nasty border at the borders of the passages. Here are these typical landscapes, when they are trying to draw out a distant very dark forest with the "shadow" tool, and above the edge of the trees there is a strip that limits the area of the mask created by the blurred image.
Well, if you want to draw out individual bright areas - lights, the sun - you can make a stack of several images - with real bracketing or like this, artificial, and separately, using handles, to show the necessary elements using a mask. Apologists for erotic forms of inaccessible ideals will begin to wrinkle their noses, but competent photo processing is sometimes a more significant tool than the most coveted camera with the most legendary lens on it.
These sliders work with the same image, so the palette will be the same. The algorithms used are different, from this and the border. HDR is a cool thing, but programs usually do not correctly glue the exposure, so I forgot about this thing - it's better to apply a mask
The original image is 12 or 14 bit, stretched to 16 bit. Target - 8-bit. Usually “tails” in highlights and shadows are cut off, there are few significant pixels. But if the picture is very contrasting, such as hard daylight in the open sun, a dark key or a light key, or a banal wedding with a groom in a dark suit and a bride in a snow-white dress (there are still options with skin colors), then the image can be compressed by pulling up the tails with one side or the other. Real HDR would give the opportunity to depict the entire tonal range, but in the absence of such an opportunity, a kind of "tone mapping" is used. At the same time, the image quality does not suffer, on the contrary, we add those details that could be lost.
But these “pseudo-HDRs” are shadows / highlights on steroids, when dark and light pixels are pulled from the tails to the center, because they are not visible at the edges, and the eye's sensitivity is lower, and the gamma curve gives maximum contrast in the center ... In this case, the pixels do not stretch all, but along the mask, which is a blurred image. Something must remain white, and something black, otherwise the image will be tonally incomplete. But it looks pretty jagged and dumb. I saw a dozen images - I saw everything.
Yes, it's all clear, okay, let's go)
I have sometimes had to use a combination of differently processed files from the same RAW. But I did not use an application for HDR photos, but manually in the GIMP editor loaded files with different processing into layers, then wiped areas in different layers and got a combined image. Here is an example of a photo (Panasonic Lumix DMC-G3 camera, 2017, now I probably would do better) :(https://ibb.co/NC8JqDp) From additional layers I added faces and also background - sky and buildings. Original JPG from camera: (https://ibb.co/7K1kM2w)
I also tried to master traditional HDR before, I used a plugin for GIMP. The results weren't great, but the original photos themselves weren't great either. Now I found another plugin for GIMP, similar to those that I used earlier. But with this, the results are much better, and my original photos are now better too. This autumn I just took pictures of buildings in Kiev with evening illumination for the competition "Vіki to love pam'yatki". It turns out, for example, like this: (https://ibb.co/LhQMXRV), original JPG from camera: (https://ibb.co/Bnj1jfQ)
EXIF shows that this is not a G3, but a G80. And it appeared in 2011, not 2017.
I have a new Panasonic Lumix DMC-G80 camera, HDR photos were taken on it. And the first photos on the G3, then I had one.
I downloaded the file from the first example again - JPG from camera (https://ibb.co/L5ttCcL), I downloaded the wrong file first.
In portraits, pseudo HDR is often redundant, but yes, it's an excellent technique, I use it all the time.
I use Lightroom. I downloaded plugins and among them there is HDR processing. The RAW file produces sharper, more vibrant photos. Is this also a pseudo-HDR? Or is it something else?
Yes, + - the same - dd extension from one RAW file