Re: So I had not really tried shooting a human yet
I don't mean to beat a dead horse but...if compositing 3 photos taken at different exposures didn't offer a different result, then why waste the disk space or processing time?
Re: So I had not really tried shooting a human yet
the middel pic Colin just posted is amazing...but not really if yah know a few things..hehe:):)
Re: So I had not really tried shooting a human yet
Quote:
Originally Posted by
DarinHughes
I don't mean to beat a dead horse but...if compositing 3 photos taken at different exposures didn't offer a different result, then why waste the disk space or processing time?
It does offer a different result in that the alternative would be an image with no shadow detail or blown highlights.
HDR is really nothing more than a set of techniques to accomplish what can't be accomplished in a single exposure; whether someone subsequently chooses to process that as something photorealistic or something more artistic is really up to them. Case in point, the first of the 3 images in my example above was from a commerial shoot for a brochure for a rest home - it wasn't possible to get that they wanted with a single exposure (even with GND filters) - and there is no way any "artistic" interpretation would have been acceptable to the client ...
... so it wasn't a waste of time or disk space, and the money was just fine thank you :)
Re: So I had not really tried shooting a human yet
Well...then maybe I didn't understand your post 'cause I think we are making the same point. A HDR composite does not look the same as a standard photograph (even though it is necessarily presented as a LDR image). And while you can make a HDR composite look completely realistic, it still has properties that you would not find in a traditional photograph. 'Cause the photos you posted are incredible and unlike anything I have ever seen in a single exposure.
My question is...so how then should we refer to these photographs?
Re: So I had not really tried shooting a human yet
Quote:
Originally Posted by
DarinHughes
Well...then maybe I didn't understand your post 'cause I think we are making the same point. A HDR composite does not look the same as a standard photograph (even though it is necessarily presented as a LDR image). And while you can make a HDR composite look completely realistic, it still has properties that you would not find in a traditional photograph. 'Cause the photos you posted are incredible and unlike anything I have ever seen in a single exposure.
My question is...so how then should we refer to these photographs?
The "properties" you refer to are probably proportional to the EV range of the original scene - at one extreme you can have something like #2 with naked sun whilst still retaining the mid-tones I wanted, to ones like #3 where, if I hadn't used a GND filter then I'd have had to dig the foreground detail out from around the noise floor ... it would probably still have done it, but they'd have been VERY noisy. The greater the EV range captured in the original scene (and desired, compressed, into the final image) the greater the challenge to get them looking "natural" - but it can be done (some so well that you wouldn't even think HDR, even though it had naked sun in it).
How should we refer to these photos? Who knows. Certainly what photographers take as being the norm is a normal dynamic range output (because we don't have anything that can readily display more than about 5 stops) - so in my mind - in a photography sense - I like to think of it as being where the dynamic range of the original scene is too great to capture with a normal single exposure (so special techniques such as bracketing or GND filters are needed to modify the dynamic range of the scene into something we can work with).
The bit I'm REALLY trying to get people away from is people mistaking a certain "highly processed look" - especially when processed from a single exposure (without GND filters) - as being "the defining characteristic of HDR". In my mind, what defines HDR (from a general photography perspective) is what happened with the capturing of the original scene, not how realistic or processed the final result looks.
Re: So I had not really tried shooting a human yet
Quote:
Originally Posted by
Colin Southern
The bit I'm REALLY trying to get people away from is people mistaking a certain "highly processed look" - especially when processed from a single exposure (without GND filters) - as being "the defining characteristic of HDR". In my mind, what defines HDR (from a general photography perspective) is what happened with the capturing of the original scene, not how realistic or processed the final result looks.
Right on. That is exactly the point that the CS student was making to me earlier today. His concern was that when people search for HDR photography they see these very stylized photos and assume HDR is just another photoshop filter and thus discount all of its capabilities.
Re: So I had not really tried shooting a human yet
Technically I too would side with your friend on the point it's not correct to call them hdr but admittedly it's a bit of a mouthful to give the full terminology. As you know in photograph you make the HDR from composite of LDR images which collectively encompass higher range than possible in 1 exposure (*) then tonemap this actual HDR image to generated a LDR. Yes, a bit of a mouthful and people who understand can implicity read all that into the phrase "from HDR (3 exposures)". This is why it's important for me to understand it. ( * = I say this because 7 exposures that collectively add up to 5 stops range is still a LDRI image that has no more info than what 1 well exposed image could contain hence most scenes range can be captured in 2 or 3 shots).
Playing devils advocate though I know a lot of the rendering and computer community get upset about some about the terminology or technical differences between HDRI application in rendering and photography but I think a lot of it is forgivable since it's quite a different process and purpose and there is little real world confusion between HDR in photography and cgi so I personally think it doesn't matter so much. I am very new to HDR in photography myself but have been familiar with it primarily for use in lightmaps for rendering for a while and it's a very different world although many things translate well a lot of it doesn't.
Also the tech side people often get a bit oversensitive about such things, much more than need be (myself included). It gets silly like the whole you cannot say "linux" anymore but have to say "GNU project blah blah blah Linux blah blah", it's nit picking and pointless in essence. There is a big difference between simply shortening something for ease of communication (which is a reasonable thing to do) and perpetuating a myth or misunderstanding I think. Finally on a nit picking tech point HDR application in photography was developed to produce images that look identical to a standard image as colin said so technically you shouldn't be able to see a difference in the final image from a "regular" one as that was the very reason it was developed for :D . Personally though I think if people want to do things differently to intended development purpose to make artistic/surreal images then there is no problem in my eyes although I think it helpful if they understand the technical truths.
Re: So I had not really tried shooting a human yet
I use tonemapping of a single image by generating -1,0,+1 ev images from a raw file to raise low tones without noise or very little; where I have found other ways tend to increase noise more.
However I haven't an example except a self portrait and this would be too frightening to show here.
The exif gets nuked though. :)