Show us your HDR photos.

So you are doing the blending manually? Photoshops HDR won't let you use a single image that has had the expsoure adjusted. Adobe says that doing so doesn't provide any new real data. Apparently what ever you do to the single image to make it lighter or darker doesn't really reveal any more information to the plug-in. We see more information, but apparently the program that does the work doesn't and so it won't let you do it with a single image JPG or RAW.

Robert
 
Personally I do the blending through the Photomatix product. I develop the raw images through RSP.

When you develop a RAW image to a jpeg, a lot of detail is discarded, the the hdr process uses that information. I agree with you it doesnt look realistic. It often ends up looking like a very detailed painting. Whether you like it or not is personal taste. I think it can work on some images. As an example, the shot below was taken on a very flat grey day. The hdr image is below:



 
So you are doing the blending manually?
Photoshops HDR won't let
you use a single image that has had the expsoure adjusted. Adobe
says that doing so doesn't provide any new real data. Apparently
what ever you do to the single image to make it lighter or darker
doesn't really reveal any more information to the plug-in. We see
more information, but apparently the program that does the work
doesn't and so it won't let you do it with a single image JPG or
RAW.
Hmm I am not sure that I understood your point... When I was talking about tripod, this is because I think that to merge pictures to HDR (with photoshop), you need at least 3 pictures (shoot with a tripod) with different exposures. Is it correct ? Or did anybody use another technic (except Raw which I can't use) ?

Thanks

Maud
--
Have fun while taking pictures !!
Please critic my pictures :-)

FZ 20 and Sunpack 383

few galleries : http://membres.lycos.fr/oiseau1/
 
No a tripod is needed so that the 3 or more images line up perfectly other wise you get ghosting and blurring.

Robert
 
So you are doing the blending manually? Photoshops HDR won't let
you use a single image that has had the expsoure adjusted. Adobe
says that doing so doesn't provide any new real data.
and that's correct - doing math to an image creates no new data. doesn't that make sense? it sure does to me.
Apparently
what ever you do to the single image to make it lighter or darker
doesn't really reveal any more information to the plug-in.
and why should it?
We see
more information, but apparently the program that does the work
doesn't and so it won't let you do it with a single image JPG or
RAW.
you don't see more info - not when there is no more info TO see.

this isn't just photo - this is 'info theory'. if you take a data set and scale it, linearly, did you get 'new' data from that? no, its DERIVED data. new data is collecting whole new data 'from the field' and not just taking same-old data and rehashing it.

if you asked the same person a question and got an answer, do you expect different results when you ask again and again? ;)

same with data. you need new pixel sets to do the averaging. averaging 5 and 5 and 5 and 5 and 5 will always yield the same old boring 5. right?

--
bryan (pics @ http://www.flickr.com/photos/linux-works )
(pics and more @ http://www.netstuff.org ) ~
 
Yes, but when you convert the RAW to a JPEG with a given exposure value, you lose a lot of data. By developing with different exposure values, you end up with multiple JPEGS, one exposed for the shadows, and one for the highlights, say. Then you can combine in the Photomatix software (or whatever) to give you the HDR and then a tone-mapped image. The tone-mapping takes the HDR image (which contains more data than you can display) and tries to even out the overall exposure, while keeping detail contrast in the picture. That's why it looks slightly unnatural I think.
 
And then we'll have to agree to disagree ;)

I know as a mathematician that you are right, there is one set of data. But this is my understanding of HDR:

When you look at a RAW file on, say RSE which is what I use, you can slide the exposure slider or the contrast sliders, to reveal or not reveal detail. The RAW file has more data than you can display. For any given combination of settings, you either lose highlights, shadows, or mid-range detail in the visible photo. The data is still there.

What HDR and tone-mapping does (Photomatix, that is, I don't know about the Photoshop version), is try to compress the data in such a way that local details are retained in shadows and highlight areas, and it does it by averaging the data at a global level, but keeping local contrast alive. So bright sky areas are darkened, on average, while shadow areas are lightened, again on average. However there is still local contrast. It's a bit like choosing different RAW exposure values for different areas of the final photo.

If you look at the two pics I posted earlier today of Newquay Harbour, the original JPEG has not much detail in the shadows of the wall, whereas these details were in the RAW file and have come out in the HDR process. The result looks unnatural, but it is real detail.

And the process itself is quite fun as well ;)
 
The human does see more. You take a dark shadow area and you lighten it so you can see what is in the shadow the human is going to see more information. That doesn't mean the information wasn't there all of the image, the computer and software saw it but the human did not.

Robert
 
Talk to Adobe about HDR. They will tell you and the plug-in in Photoshop CS2 will not allow you to use three seperate images (with different exposures done in ACR) for an HDR image. The plug-in knows you are trying to use one image for the three and will not allow you to do it.

Adobe has covered this very well. It doesn't work and can't be done. Now if other software is allowing you to do this. It means either the software is less picky and isn't doing the job right and/or the software isn't very good and can't tell that you are using the same image.

Personally, I will go with Adobe. It doesn't matter what you do to a raw file as far as adjusting the exposure. The computer, the software and the HDR plug-in sees all data in the file even if the human can't. When you adjust exposure in an image you are adjusting for the benefit of the humans looking at it and not the software or the hardware. Just for the humans.

Take a dark shadow area. The computer and software know what is in there. When you adjust it so you can see it, that is only for the benefit of the human, not the software the software and the HDR plug-in has always saw that data. Plain and simple.

Robert
 
...(half devoured mushroom)
 
Talk to Adobe about HDR. They will tell you and the plug-in in
Photoshop CS2 will not allow you to use three seperate images (with
different exposures done in ACR) for an HDR image. The plug-in
knows you are trying to use one image for the three and will not
allow you to do it.
Just because the software doesn't allow you to do something
doesn't mean there is no value in combining different 'developments'
of the same RAW file. Not allowing this is merely some software developer's
decision as to what they will allow their software to do. Do you
always follow the guidelines set down by authority figures?
Sounds so conformist.
Adobe has covered this very well. It doesn't work and can't be
done. Now if other software is allowing you to do this. It means
either the software is less picky and isn't doing the job right
and/or the software isn't very good and can't tell that you are
using the same image.
Who said it doesn't work? I do it all the time, with results that
I can plainly see, and which improve my images when they have
exposure issues. I am doing the job with Photoshop and ACR,
but not the HDR plugin. There are tutorials I could point you to
on the web explaining the technique of combining multiple
conversions of a given RAW file.

Part of the point here is that some cameras (DSLR's for sure,
I can't vouch for P&S's since I don't have one with RAW
output capability) have a wider gamut (larger color space)
than sRGB. The human eye can see more than sRGB jpeg
can encode in terms of color. One place this is quite apparent
is in the tendency of jpegs to have blown reds, a subject much
discussed here a while ago. Using a larger color space such
as Adobe or Prophoto RGB allows one to capture a wider gamut,
and using two different developments allows one to compress
the highlights and shadows of that color into the smaller
sRGB color space. I recently had a bright orange flower
with a hummingbird feeding on it.
In sRGB the flower was completely overexposed
if I used sRGB and kept the bird from being totally dark.
Using Prophoto RGB gamut allowed me to keep tonal range in
the flower without underexposing the bird. Alternatively, I could
use two RAW conversions, one for the flower and one for the bird.

The same holds true for overall luminance values. I often use
two RAW conversions of the same file and combine them
in layers, using them to lighten the shadows and/or bring back
the highlight details.
Personally, I will go with Adobe. It doesn't matter what you do to
a raw file as far as adjusting the exposure. The computer, the
software and the HDR plug-in sees all data in the file even if the
human can't. When you adjust exposure in an image you are adjusting
for the benefit of the humans looking at it and not the software or
the hardware. Just for the humans.

Take a dark shadow area. The computer and software know what is in
there. When you adjust it so you can see it, that is only for the
benefit of the human, not the software the software and the HDR
plug-in has always saw that data. Plain and simple.
You seem to have an overly restrictive conception of what the RAW
data represents. It is as the name indicates the unprocessed data
captured by the sensor. Any conversion of that data to another format
inevitably involves processing of the data. Vanilla conversion to
sRGB is not guaranteed to match ideally the sensor data, in fact
generically it clips in both the shadows and highlights
of some colors. One can use two conversions of the same RAW file
to bring the image data within the more restrictive confines
of the 8-bit sRGB gamut of a jpeg image.
I don't see why this amounts to fakery; it is
merely compressing the data to fit within the confines of an
overly restrictive standard of what digital color and exposure
should be, since indeed the human eye has a wider gamut than
sRGB, and one would like to represent the tonal range in
some way that our eyes can interpret and appreciate.
Granted, this may be less of an issue with P&S cameras
with their more limited dynamic range. But with DSLR's
it is definitely a useful tool, even if the people who developed
the HDR plugin couldn't think of a way to incorporate it
in their software.
--
emil
--



http://theory.uchicago.edu/~ejm/pix/20d/
 
I never said there was no value in it. What I am saying is that it isn't a real HDR image. And, it isn't.

Robert
 
Alright I think the problem here is one of definitions. Adobes definition of HDR seems different, in that it's an averaging or compressing technique, and I agree, if you follow that definition, you don't gain anything.

But the Photomatix product (and there may be others I don't know), is doing something in addition. It averages over the whole area of the photo, but retains contrast locally, i.e. in the small details. This means that you can get detail in both the shadow and highlight areas in the same jpeg assuming the detail is in the RAW file to start with.

The Shadow areas become lighter, on average, and the lighter areas become darker.

If you look at this one:



The only way I could get the detail in the houses, and the detail in the sky in the same jpeg was by using Photomatix HDR and tone mapping basedon 2 different developments of the same RAW file. There was no setting that allowed me to get that from the RAW file using my RAW developer. Photomatix has 'fiddled' witht e light and shade areas to bring them both to the same average sort of level, but keeping the contrast in the details.

You may not like the results, but it does work.

--
watching-the-time.blogtog.com
 
Hi all,
Recently I started experimenting with Photomatix.
Hi Voe,

What are Photomatix and HDR photos ?

JJDiniz, from Brazil

.
 
JJDiniz wrote:

HDR stands for High Dynamic Range, it's a process whereby one uses multiexposed images and blends them into one final image. Thus achieve more detail eliminating the dark and overexposed areas.
Photomatix is a software tool that you can use to do that.
Hi all,
Recently I started experimenting with Photomatix.
Hi Voe,

What are Photomatix and HDR photos ?

JJDiniz, from Brazil

.
--
http://www.flickr.com/photos/voe/
 

Keyboard shortcuts

Back
Top