Pages: 1 2 3 [4] 5 6   Go Down

Author Topic: Uwe Steinmuller of DOP on dynamic range and HDR  (Read 53162 times)

PierreVandevenne

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 512
    • http://www.datarescue.com/life

reflectivity is only about 50:1, or under 6 stops.

Prints don't look like the ideal media to demonstrate a significant DR difference between DSLRs and large format cameras then...

PS: sorry, couldn't resist.  ::)
Logged

NikoJorj

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1082
    • http://nikojorj.free.fr/
Re: Uwe Steinmuller of DOP on dynamic range and HDR
« Reply #61 on: December 30, 2010, 04:55:27 pm »

It's pretty common to have tonal inversions in these types of images, [...]
One may find some tonal inversions (or shifts at least) in human vision too, see the well-known checkerboard :



I do believe some slight haloing in tonemapping is unavoidable with nowadays tools, the thing is rather to hide the dust under the carpet (french proverbial to say that it should go under the radar).
Logged
Nicolas from Grenoble
A small gallery

RFPhotography

  • Guest
Re: About bracketing for HDR and output devices DR
« Reply #62 on: December 30, 2010, 08:58:31 pm »


I'm not sure everyone realizes that the HDR technique involves a move into the space of absolute magnitudes, and away from relative white-black point of a single capture.  This is a conceptual shift.  I think some here are carrying over the assumption that HDR is just another tool for doing LDR, but the conceptual shift is more significant.

True enough, Luke.  But I think a big part of the reason that realisation is lacking is because, right now, the technology of the image is far ahead of the technology of the presentation.  The fact that we have to come back to an LDR space to work with and view these images takes away the advantage of; to some extent, and slows down the full understanding of the benefits of the larger bit space.  When the presentation technology - in particular monitors - catches up there's going to be a big 'Wow!' moment.   :)

Feppe, leave it to Keke to come up with an interesting quote.

Something else that doesn't work in all of this is the idea that bit depth and dynamic range are interdependent.  People are combining the concepts where no combination is required.  The two are not interdependent.  A higher bit depth does not, in and of itself, mean a higher dynamic range.  Bit depth simply means there are more in between tones from dmax to dmin.  So while it is true that you don't need to leave the low bit depth environment to increase dynamic range, moving into the high bit depth space (and floating point) has distinct advantages when you start trying to move those pixels around and particularly when you need to push the pixels around significantly to get them back into the kiddie-sized LDR pool. 

Sure, if sensors could capture 16 or 18 stops of brightness with absolute perfection in real world (as opposed to the lab) conditions it may reduce the need (although not eliminate) for HDR.  But if ifs and buts were candies and nuts we'd all have a Merry Christmas too.  The fact is cameras can't do that and while some may say it's inevitable - and it may be - my bet is it won't happen in the next 5 years so until then we use the tools we have at hand to the best of our abilities.
Logged

Guillermo Luijk

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2005
    • http://www.guillermoluijk.com
Re: About bracketing for HDR and output devices DR
« Reply #63 on: December 30, 2010, 09:42:34 pm »

Sure, if sensors could capture 16 or 18 stops of brightness with absolute perfection in real world (as opposed to the lab) conditions it may reduce the need (although not eliminate) for HDR.  But if ifs and buts were candies and nuts we'd all have a Merry Christmas too.  The fact is cameras can't do that and while some may say it's inevitable - and it may be - my bet is it won't happen in the next 5 years so until then we use the tools we have at hand to the best of our abilities.

The journey will be progressive, it won't have a clear deadline. And this journey is already in progress. You don't need a 16-stops DR sensor to capture a 12 stops scene in a single shot. But you need to bracket a 12 stops scene if all you have is a 8-stops DR camera designed 5 years ago.

I shot this 12 stops scene in the Summer of 2007 with my Canon 350D (8 stops effective DR). Of course I needed to bracket {0, +4}:




Today's Pentax K5 effective DR is 11 stops. Translating its 16 Mpx into a 38Mpx FF sensor using the same photosites (i.e. already existing technology), we get a sensor with about 12 stops effective DR for the same output resolution.

What 3 years ago I needed to bracket 4 stops, could be captured using today's technology in a single shot (just wait for the next FF cameras appearing). The same will happen tomorrow with 14 stops scenes, and the day after tomorrow with 16 stops scenes, and so on until 99% of the real world scenes don't deserve any bracketing. Slowly, but surely.

Regards
« Last Edit: December 31, 2010, 06:25:38 am by Guillermo Luijk »
Logged

John Camp

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2171
Re: Uwe Steinmuller of DOP on dynamic range and HDR
« Reply #64 on: December 30, 2010, 10:48:04 pm »

I dont know him. Does his images look like they were taken at any random place at any random time, or does it look like he has carefully chose time, place and camera settings to make a visually pleasing image?

JC: Who knows? And what difference would that make?

If he in any way is "putting his soul" into his image, I would say that that could detract from the realism but add to the artistic value.

JC: It could detract from the realism, but add to the artistic value, but then again, maybe not. A person with an inane vision could put his soul into a work and have it come out...inane.

BTW, do you think that art should be valued from the end-result alone or does knowledge of the process add/subtract to its value?

JC: Could be either one.

If I show you an amazing image that blows your socks off (purely hypothetically speaking), would you be any less impressed if I told you I had made it purely in Photoshop?

JC: Probably. But that's just me. Other people might regard it as great art.

Or is the ideal that one should wait for weeks in a cold, deserted place waiting for "just the right light" and then capture that magic moment right before the batteries run out and being tragically eaten by a bear?

JC: I don't think there is an ideal.

You are mixing arguments here. If "lack of realism" is a valid argument against some art it should be a valid argument against all art.

JC: Really? If an argument against one woman is valid, is that an argument against all women? Frankly, this suggestion makes no sense at all. I'd heap further ridicule on it, but that that would take too much time.

If the critique is that it "is not suceeding in its own terms", then that it the argument that you should use.

JC: That is more or less the argument that I use, except that even if it does succeed in its own terms, it may not be art. My cat snapshots succeed in their own terms, but they remain cat snapshots. But if someone takes a stab an producing art, and the effort fails in its own terms, then it probably isn't high art.

You second statement statement seems irrelevant to what I said.

JC: I would disagree.

-h
« Last Edit: December 30, 2010, 10:55:26 pm by John Camp »
Logged

John Camp

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2171
Re: About bracketing for HDR and output devices DR
« Reply #65 on: December 30, 2010, 10:52:38 pm »

John, you've made the same mistake as many others.  You've done it with respect to art in general as opposed to the ones who address HDR specifically.  You've imparted your objective position onto a subjective subject.  And that is what's wrong.  And that's not a subjective issue.  HJ has suggested what he 'expects' art to be.  An expectation isn't a hard and fast, objective construct.  Anything that captures or freezes a moment in time isn't realistic.  If I can't go to that place and see exactly what is in that photo or painting or movie or drawing or 3D rendering then it's not realistic.  The only true realism is what I, or anyone else, can see with my own eyes.  I can choose to believe or not the reality someone else saw and the way they present that reality to me and accept it as real but it's not truly real to me. 

I don't think you read what I said carefully enough, because if I understand what you're saying, we're more or less off in the same direction.
Logged

Ray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 10365
Re: Uwe Steinmuller of DOP on dynamic range and HDR
« Reply #66 on: December 30, 2010, 10:53:02 pm »

One may find some tonal inversions (or shifts at least) in human vision too, see the well-known checkerboard :



The implications of this effect shown in the checkerboard, both squares A and B having mathematically the same brightness level (107, 107, 107), are mindboggling.

This effect, to varying degrees, will apply across the whole tonal range in any image, including mathematically identical color hues appearing visually quite different according to their context within the image.

I generally do not use specific programs that are described as tone mapping programs, but in practice I usually tone-map my images to get a result which I like, using the Shadows/Highlights tool, Brightness/Contrast tool, or simply selecting areas with the lasso tool, feathering significantly, then making whatever adjustments to color, tone and brightness I think appropriate.

I'm no wizard at using Photoshop, but one technique I use often to make adjustments whilst simultaneously protecting the highlights is, 'Ctrl + left click' on RGB Channels, invert the selection, go to Layers/New Adjustment Layer,  select whatever tool is appropriate (for example, Levels), set opacity to say, 80%, then make the adjustment.

This method allows one to brighten the entire image, whilst maintaining full detail in the highlights.  I find it very useful. I learnt this technique on Luminous Landscape.

Of course, if the darker parts of the image can't withstand brightening without revealing ugly noise, then you're stuffed  ;D .
Logged

Schewe

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6229
    • http:www.schewephoto.com
Re: Uwe Steinmuller of DOP on dynamic range and HDR
« Reply #67 on: December 30, 2010, 11:23:43 pm »

If some people like the stylized HDR look with aggressive "detail enhancement" that's fine. Different people have different tastes; and when it comes to art anything goes, so I certainly don't think that a naturalistic approach is the only valid one. I can appreciate truly well-done stylized HDR, even if it's not to my personal taste. The problem is, it's extremely rare. The vast majority of stylized HDR imagery is full of ugly artifacts that I just can't see past, and it boggles my mind that so many people don't seem to mind the ugly halos, color shifts, etc. Hopefully over time the tools will get better and this will improve; but right now I would say that the "bad" HDR outweighs the good by at least 10:1. So for a lot of people, this pretty much spoils the whole genre.

Which was my original point regarding Uwe advocating and teaching HDR images that look surreal (which is how I respond rather than saying "stylized" which some how kinda lets people off the hook).

Compressing a high contrast scene into a printable dynamic range is indeed difficult. But it can be done without all the surreal downside. I would encourage people to actually learn how to do it so it isn't glaringly obvious. Which I'm not sure Uwe's tutorial does...
Logged

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Re: Uwe Steinmuller of DOP on dynamic range and HDR
« Reply #68 on: December 31, 2010, 12:36:45 am »

Hi,

Here is another try. Three exposures combined and than using LR-controlls for final image.

http://echophoto.smugmug.com/Special-methods/HDR/HDR/20101214-DSC09731/1142183148_uNBtR-X2.jpg

Best regards
Erik



Which was my original point regarding Uwe advocating and teaching HDR images that look surreal (which is how I respond rather than saying "stylized" which some how kinda lets people off the hook).

Compressing a high contrast scene into a printable dynamic range is indeed difficult. But it can be done without all the surreal downside. I would encourage people to actually learn how to do it so it isn't glaringly obvious. Which I'm not sure Uwe's tutorial does...
Logged
Erik Kaffehr
 

hjulenissen

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2051
Re: Uwe Steinmuller of DOP on dynamic range and HDR
« Reply #69 on: December 31, 2010, 02:15:06 am »

Actually I think you can argue mathematically that the heavily stylized HDR look tends to be less realistic. It's pretty common to have tonal inversions in these types of images, where for instance the shadowed foreground is actually brighter than the daytime sky, just to name one very common example. So I don't really think you can argue that folks think this stuff looks unnatural just because film came first.  Maybe if the real world looked like the one in Avatar this argument might hold some water...
There are errors in tonemapped images, yes. Do you think that blown-out highlights and clipped blacks are a part of what you normally see in a scene? So there are errors in regular images as well. I do not see any attempt at bringing out mathematical tools to support your statement?

-h
Logged

hjulenissen

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2051
Re: Uwe Steinmuller of DOP on dynamic range and HDR
« Reply #70 on: December 31, 2010, 02:22:06 am »

Your way of quoting makes it very hard for readers and repliers.

Quote
JC: Who knows? And what difference would that make?
I am supporting my claim that I expect art to be something different than taking a snapshot of reality. That was why you started this discussion with me in the first place, was it not?
Quote
Quote
You are mixing arguments here. If "lack of realism" is a valid argument against some art it should be a valid argument against all art.
JC: Really? If an argument against one woman is valid, is that an argument against all women? Frankly, this suggestion makes no sense at all. I'd heap further ridicule on it, but that that would take too much time.
If you are saying that "HDR is crap because it is not realistic", then you are saying that not being realistic makes it crap. If you at another stage claim that some other imagery is great even though it is not realistic, then you are not honest in your arguments.

If a rule is universal, it must be valid everywhere. If it is not universal, it should carry a disclaimer. DR sucks because it is unrealistic, and I dont lik"
Quote
Quote
You second statement statement seems irrelevant to what I said.
JC: I would disagree.

That is your right. If you wont bother relating your statements to mine, I wont bother replying.
-h
« Last Edit: December 31, 2010, 02:50:57 am by hjulenissen »
Logged

stever

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1250
Re: Uwe Steinmuller of DOP on dynamic range and HDR
« Reply #71 on: December 31, 2010, 09:25:11 am »

thanks for the encouragement Schewe, (i do believe it can be done) - does anyone have a tutorial have a tutorial that teaches how it's done?
Logged

RFPhotography

  • Guest
Re: Uwe Steinmuller of DOP on dynamic range and HDR
« Reply #72 on: December 31, 2010, 11:30:13 am »

It's not possible, Steve.  The reason it's not possible is because every image is different.  There are no 'set smoothing to 25, brightness to 50, saturation to 30, etc.' formulae for more realistic images.  There's a learning curve involved.  It's also not possible because every software app. is different.  Time needs to be spent learning the software, how it works, what the various tonemapping operators do, how they work independently and how each affects the others in combination. 

It is possible to make general statements about how different operators impact an image and within those general statements one can get an idea of where to start to get a more 'natural' result.  But that's really as far as it can go.  That's what I've done in my HDR tutorial.  I also have three presets people can download and use that offer starting points for three different 'looks' - a slightly unreal, sort of graphic illustration look, a natural look and a hyper-grunge look. 
Logged

Ray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 10365
Re: Uwe Steinmuller of DOP on dynamic range and HDR
« Reply #73 on: December 31, 2010, 08:11:39 pm »

It's not possible, Steve.  The reason it's not possible is because every image is different.  There are no 'set smoothing to 25, brightness to 50, saturation to 30, etc.' formulae for more realistic images.  There's a learning curve involved.  It's also not possible because every software app. is different.  Time needs to be spent learning the software, how it works, what the various tonemapping operators do, how they work independently and how each affects the others in combination. 

It is possible to make general statements about how different operators impact an image and within those general statements one can get an idea of where to start to get a more 'natural' result.  But that's really as far as it can go.  That's what I've done in my HDR tutorial.  I also have three presets people can download and use that offer starting points for three different 'looks' - a slightly unreal, sort of graphic illustration look, a natural look and a hyper-grunge look. 


I compeletely agree with Bob here.  If you are looking for a general formula that can be applied to make an image look natural, then you might as well just shoot in jpeg mode and let the camera apply its own built-in adjustments.

A number of different exposures which have been merged to HDR, becomes a single image which needs to be adjusted as any single RAW image needs to be adjusted during and after conversion. It's rare that an image can look exactly right with just a click on the 'auto' button in ACR. If it does, it will still need further adjustment in 'proof mode' before printing.

If the result doesn't look satisfactory, for whatever reason, then the photographer is to blame (or the person who processed the image). Don't blame the tool. Photoshop is an amazing tool for image adjustment.
Logged

LKaven

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1060
Re: Uwe Steinmuller of DOP on dynamic range and HDR
« Reply #74 on: January 01, 2011, 01:09:14 am »


I compeletely agree with Bob here.  If you are looking for a general formula that can be applied to make an image look natural, then you might as well just shoot in jpeg mode and let the camera apply its own built-in adjustments.

A number of different exposures which have been merged to HDR, becomes a single image which needs to be adjusted as any single RAW image needs to be adjusted during and after conversion. It's rare that an image can look exactly right with just a click on the 'auto' button in ACR. If it does, it will still need further adjustment in 'proof mode' before printing.

If the result doesn't look satisfactory, for whatever reason, then the photographer is to blame (or the person who processed the image). Don't blame the tool. Photoshop is an amazing tool for image adjustment.
All true and more.  It's a little like re-lighting and re-taking the picture. 

And the common tools have arcane user interfaces.  Many common settings are calibrated according to a "0-100" scale, which are simple to remember mentally, but give no indication of the underlying variables being manipulated, nor what their true values are.  The combinatorial possibilities of the controls are too many, and to get a good image, you really have to learn to either play it like an instrument, or to use what you get in the most honestly artistic way possible.

LKaven

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1060
Re: About bracketing for HDR and output devices DR
« Reply #75 on: January 01, 2011, 01:48:05 am »

A bit dissapointed to read this:

It makes me think Steinmueller didn't really get that the point of bracketing for HDR will soon be unnecessary, and it does not participate in the definition of HDR itself. The only reason we have today for bracketing HDR scenes is that sensors are still too noisy to capture in a single shot the entire DR of many real world scenes.
Guillermo!  I'm surprised to hear this from you, as a man who is as concerned about signal optimization.

What we are talking about here is a broader concept.  Imagine just in one part, the idea of allocating bits /where the information is/ as a central concept.  Think of that in terms of this one practical case.

Recently I did a portrait of someone in a church space, where the subject was lighted by a stained glass window.  The inside of the church was dim, but beautiful.  Now even with a D3x, a camera with good dynamic range at ISO 100, I could not capture any detail whatsoever on the inside of the church.  It simply came out mostly RGB=0,0,0.  There were a few single bit quantities, but nothing discernable.  

It is important to ask here - why must the church be black?  It doesn't look black to me.  But it is arbitrarily the case, partly by virtue of the original physical chemistry employed, that each individual exposure has an implicit black point and an implicit white point, both of which are /false/.  

What I could have done was (1) shoot the background in HDR, (2) do portrait takes, and (3) composite them.  But this would just be a way of allocating bits to the relevant content.   Audio encoding schemes allocate bits where the psychologically salient information lies.  Photography should record visual stimuli where the salient information lies -- in absolute magnitude space.  

By supersampling the scene inside the dim church, I could have collected that information.  We're no more obliged than any painter to make the dynamic range of a source image correspond to the dynamic range of the output medium.  The sun can burn orange, as viewed from the interior of a candlelit chamber through the window onto a blue sky peppered with cirrus clouds.  And when you paint the candlelit interior, it will be detailed.    Don't we, with our inherently diachronic visual system, kind of see it this way?

Now finally, imagine this.  A camera to be could be capable of flexibly and adaptively supersampling a scene by allocating collection of information, locally as well as globally, over a given shutter interval.  Under its control could be differential gain between pixels, and localized multiple exposure.  Normalization and averaging could be done in camera.  The reason this would have to be done in camera is because in practical terms, you are interested in events that last for only about 1/60th of a second.  In order to carry out a complex "superexposure" program, you'd have to hand the task over to the camera.  

I really believe this is coming.  And this tells you something of why I think this is more than a fad involving cheesy special effects.
« Last Edit: January 01, 2011, 02:11:21 am by LKaven »
Logged

HCHeyerdahl

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 73
    • http://
Re: About bracketing for HDR and output devices DR
« Reply #76 on: January 02, 2011, 02:50:51 pm »

I'm not sure everyone realizes that the HDR technique involves a move into the space of absolute magnitudes, and away from relative white-black point of a single capture.  This is a conceptual shift.  I think some here are carrying over the assumption that HDR is just another tool for doing LDR, but the conceptual shift is more significant.

This is an interesting thread and I would like to understand this potential conceptual shift.
Up til now I thought the HDR-space was just a much larger space than a LDR-space (and hence the need for tonemapping), but in case I am missing something important, is it possible to explain a bit about this space of absolute magnitudes compared to a realtive black-white point?

Christopher
Logged

LKaven

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1060
Re: About bracketing for HDR and output devices DR
« Reply #77 on: January 03, 2011, 12:24:01 am »

This is an interesting thread and I would like to understand this potential conceptual shift.
Up til now I thought the HDR-space was just a much larger space than a LDR-space (and hence the need for tonemapping), but in case I am missing something important, is it possible to explain a bit about this space of absolute magnitudes compared to a realtive black-white point?
The HDR file is a kind of special case.  At first glance, it's just a TIF file with 32 bits.  But the data represent something other than pixels.  Think of this as a dataset of measurements, out to a good number of decimal places. 

I might characterize this space as a space of /fixed/ magnitudes as opposed to /relative/.  It might be better said that the range of magnitudes represented is literally astronomical and practically unconstrained, encompassing within one scene all possible photographic subjects in their various illuminations--fancifully, from the black cat in the coal mine, to the plasma beach on the surface of the sun--and with fine gradation.

The dataset of measurements is independent of rendering intent.  The rendering method is left open to creative and technical choice.  Methods will be refined and new ones invented.  The various "tonemapping" concepts are just a first approximation.

As a practical benefit, imagine if the 1,2,3-bit quantities in your current single-shot captures were 32-bit quantities resolved out to N decimal places?  It gives your tools something to dig into.  You could do fine-grained work on parts of your picture with high fidelity, then map them to the tonality you want.  You could even make significant changes in apparent lighting.  You could revisualize a low-key shot into a high key shot or vice-versa, selectively, with no apparent loss in image quality.

hjulenissen

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2051
Re: About bracketing for HDR and output devices DR
« Reply #78 on: January 03, 2011, 02:44:31 am »

So one can postpone exposure choices until the files are one your hard-drive. More generally, if the scene/camera allows (strictly static scenes), a 'truer' representation of the scene can be captured, and choices about exposure, black/white clipping, tonecurve etc that used to be choices of camera settings (and camera algorithms/film behaviour) can be freely chosen in the editing process of HDR. Some of these processing options may be for artistic reasons, but I would argue that most of them are to shoehorn something visually pleasing into the limited paper/display tech that we have today. We dont know what we will have in 10 years, but todays HDR pics may look better then.

I still dont think that HDR is usually 'absolute' in the sense that it is easily expressed in lumens or candela or whatever. It is still usually some arbitrary unit, but relative to that unit, all measurements are linear (and black/white clipping can be practically avoided)?

-h
Logged

HCHeyerdahl

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 73
    • http://
Re: Uwe Steinmuller of DOP on dynamic range and HDR
« Reply #79 on: January 03, 2011, 04:40:22 am »

Aha!  Thanks to both,this was clarifying :-)

Christopher
Logged
Pages: 1 2 3 [4] 5 6   Go Up