POV-Ray : Newsgroups : povray.binaries.images : More HDRI tests (136 kbu) Server Time
14 Aug 2024 05:15:26 EDT (-0400)
  More HDRI tests (136 kbu) (Message 4 to 13 of 13)  
<<< Previous 3 Messages Goto Initial 10 Messages
From: ZSpider
Subject: Re: More HDRI tests (136 kbu)
Date: 26 Dec 2002 09:29:13
Message: <3e0b1239$1@news.povray.org>
Gilles Tran wrote:
> Last ones before it gets boring... In fact I guess that similar results
> could be obtained with regular maps, lol.
> Xfrog rose, Rhino glasses. I had to touch up some residual focal blur
> graininess.

*********
Beautiful images, Gilles.  I looked at some of the HDRI links you
provided for Slime, and I understood the explanation for a while,
but then it got sorta fuzzy.  The HDR shop is a post-processor that's
applied after rendering, isn't it?  Is that what you used?

Thanks, miker


Post a reply to this message

From: Gilles Tran
Subject: Re: More HDRI tests (136 kbu)
Date: 26 Dec 2002 11:07:50
Message: <3e0b2956$1@news.povray.org>

3e0b1239$1@news.povray.org...
>Beautiful images, Gilles.
Thanks !

> The HDR shop is a post-processor that's
> applied after rendering, isn't it?  Is that what you used?

I'd rather let the experts explain what this is all about because it's quite
fuzzy for me too !
Basically the trick is to get a HDR file and use it on a big sphere in a
radiosity scene. This I figured out.
What HDR shop does is 1) let you view and edit HDR files and 2) create HDR
files out of sequences of normal images (particularly ones from digital
cameras).
Actually this part is *** very *** confusing to me right now. Anyone knows
if it possible to take a single regular image and turn it into a HDR version
? I've tried it and get a HDR file but I'm not sure it really works (or make
sense at all ?)...

G.


--
**********************
http://www.oyonale.com
**********************
- Graphic experiments
- POV-Ray and Poser computer images
- Posters


Post a reply to this message

From: Martial
Subject: Re: More HDRI tests (136 kbu)
Date: 26 Dec 2002 11:21:29
Message: <3e0b2c89$1@news.povray.org>
If you no see on  p.g or p.u.p
There are a litle bug with material_map 
The files corrected (source and exe) : 
(same name of version mlpov081 122302) 
 http://martial.rameaux.free.fr/mael/mlpov81eng.html 


Thanks Mael !

--
Martial


Post a reply to this message

From: Ive
Subject: Re: More HDRI tests (136 kbu)
Date: 26 Dec 2002 14:29:59
Message: <3e0b58b7$1@news.povray.org>
> I'd rather let the experts explain what this is all about because it's quite
> fuzzy for me too !
> Basically the trick is to get a HDR file and use it on a big sphere in a
> radiosity scene. This I figured out.
> What HDR shop does is 1) let you view and edit HDR files and 2) create HDR
> files out of sequences of normal images (particularly ones from digital
> cameras).
> Actually this part is *** very *** confusing to me right now. Anyone knows
> if it possible to take a single regular image and turn it into a HDR version
> ? I've tried it and get a HDR file but I'm not sure it really works (or make
> sense at all ?)...
>

First HDR has nothing to do with the kind of mapping you use. If you look at
Paul Debevec' page you'll find also box-mapped HDR images or simple planar
ones. High Dynamic Range Images means that the full color indensity (or better
the lumens) is also stored into the image file. This is a much better representation
of "reality" as Low Dynamic Range as represented by using the (poor) RGB
color space (used within jpg or png files). POV does internal work with high
dynamic range and I guess most of you have already used a light source that is
"brighter" than rgb <1,1,1>. I usually do so and this is the way Jaime's lighting
system does work. This is also "HDR".
Imaging a white sheet of paper, white clauds on a sunny day, the sun itself or
a exploding super nova. In rgb color space, all of them are represented as
simple rgb <1,1,1>. And as POV does already calculate with HDR values
it seems really logic and useful to me not to be restricted when it comes to
input AND output if image formats. This would give you simply the similar
possibilities that are normal to any photographer.
Oh, and one more thing: HDR means simply that the full range of gamut and
lumens is coverd by the image format. The files from Paul Debevec' page are
in the Radiance-format as developed by Greg Ward, who did also design the
LogLUV format  (also HDR). There are others around as e.g. TIFF-Pixar-Log
as used by (you guessed it) Pixar.

-Ive


P.S. sorry Gilles it did happen again, wrong button in Outlook.


Post a reply to this message

From: Slime
Subject: Re: More HDRI tests (136 kbu)
Date: 26 Dec 2002 15:20:13
Message: <3e0b647d@news.povray.org>
> Actually this part is *** very *** confusing to me right now. Anyone knows
> if it possible to take a single regular image and turn it into a HDR
version
> ? I've tried it and get a HDR file but I'm not sure it really works (or
make
> sense at all ?)...


From what I read in the links you gave me, it seems that HDR is an image
format that lets you store non-clipped color values. (normal images cover
RGB values 0-255 (really 0.0 - 1.0), whereas HDR lets you cover 0.0 to
infinity... or maybe even negative infinity to infinity, I'm not sure.)

So you couldn't get an HDR image from a single photograph, since all that
photograph contains is values from 0 to 1, and you need more photographs
with different exposure times to get more information.

That's about all I can figure out.

 - Slime
[ http://www.slimeland.com/ ]


Post a reply to this message

From: George Pantazopoulos
Subject: Re: More HDRI tests (136 kbu)
Date: 26 Dec 2002 16:55:13
Message: <3e0b7ac1@news.povray.org>
Thats correct Slime :) I've made a few square ones myself using just my
webcam and adjusting the exposure settings.. and using HDRShop.. its really
cool

George Pantzopoulos
>
>
> From what I read in the links you gave me, it seems that HDR is an image
> format that lets you store non-clipped color values. (normal images cover
> RGB values 0-255 (really 0.0 - 1.0), whereas HDR lets you cover 0.0 to
> infinity... or maybe even negative infinity to infinity, I'm not sure.)
>
> So you couldn't get an HDR image from a single photograph, since all that
> photograph contains is values from 0 to 1, and you need more photographs
> with different exposure times to get more information.
>
> That's about all I can figure out.
>
>  - Slime
> [ http://www.slimeland.com/ ]
>
>


Post a reply to this message

From: Apache
Subject: Re: More HDRI tests (136 kbu)
Date: 26 Dec 2002 17:41:44
Message: <3e0b85a8@news.povray.org>
Well.... why don't you render us some nice pics then?


:-)

Apache


Post a reply to this message

From: Gilles Tran
Subject: Re: More HDRI tests (136 kbu)
Date: 26 Dec 2002 18:33:44
Message: <3e0b91d8$1@news.povray.org>

3e0b58b7$1@news.povray.org...

> POV does internal work with high
> dynamic range and I guess most of you have already used a light source
that is
> "brighter" than rgb <1,1,1>. I usually do so and this is the way Jaime's
lighting
> system does work. This is also "HDR".

In fact my question was more, given a normal LDR image with variation of
intensity clipped between 0 and 1, is it possible to create directly a HDR
image by calculating the intensity value from the clipped one, by using a
function I=f(r,g,b) with I being maximum (and much higher than 1 of course)
for White? As I said, HDR shop accepts to do this but I'm not sure that the
final image really has hot spots (but then I have not tested this
extensively).
In any case, creating HDR maps within POV-Ray would be sweet.

G.

--
**********************
http://www.oyonale.com
**********************
- Graphic experiments
- POV-Ray and Poser computer images
- Posters


Post a reply to this message

From: Ive
Subject: Re: More HDRI tests (136 kbu)
Date: 26 Dec 2002 19:35:50
Message: <3e0ba066$1@news.povray.org>
> In fact my question was more, given a normal LDR image with variation of
> intensity clipped between 0 and 1, is it possible to create directly a HDR
> image by calculating the intensity value from the clipped one, by using a
> function I=f(r,g,b) with I being maximum (and much higher than 1 of course)
> for White?

No. If the data is clipped (like POV output) how should it be recovered. If
the data is compressed (like done with Kari's patch) there is no need to use
HDR shop, simply usse a higher ambient value, but the quality will not be the
same as with "real" hdr data. .

>As I said, HDR shop accepts to do this but I'm not sure that the
> final image really has hot spots (but then I have not tested this
> extensively).

To recreate HDR data you'll need a couple of images (lets say 5 to 10
more are better) with different exposer, this works fine.

-Ive


Post a reply to this message

From: Slime
Subject: Re: More HDRI tests (136 kbu)
Date: 26 Dec 2002 23:03:08
Message: <3e0bd0fc$1@news.povray.org>
> In fact my question was more, given a normal LDR image with variation of
> intensity clipped between 0 and 1, is it possible to create directly a HDR
> image by calculating the intensity value from the clipped one, by using a
> function I=f(r,g,b) with I being maximum (and much higher than 1 of
course)
> for White?

From a single image, you could only approximate it. All of the clipped areas
(large areas of pure white or pure black) would have to be guessed about.

 - Slime
[ http://www.slimeland.com/ ]


Post a reply to this message

<<< Previous 3 Messages Goto Initial 10 Messages

Copyright 2003-2023 Persistence of Vision Raytracer Pty. Ltd.