|
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Hi,
I got this comment regarding one of my images 'The Chess Game' which can be
seen in the images newsgroup:
"so, did you render an 8bit image out of povray?
Ideally you'd render a float precision image so you can get all the
out-of-focus hilight shapes just as bright as the in-focus hilights."
I have no idea what this means or how I do it. Any ideas?
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
the_ajj <nomail@nomail> wrote:
> "so, did you render an 8bit image out of povray?
> Ideally you'd render a float precision image so you can get all the
> out-of-focus hilight shapes just as bright as the in-focus hilights."
> I have no idea what this means or how I do it. Any ideas?
I don't understand that quote either, but my bet is that it's somehow
related to HDRI. Try googling for that acronym.
--
- Warp
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"the_ajj" <nomail@nomail> wrote in message
news:web.43749ec0ba58206ca43155f60@news.povray.org
> Hi,
>
> I got this comment regarding one of my images 'The Chess Game'
> which can be seen in the images newsgroup:
>
> "so, did you render an 8bit image out of povray?
> Ideally you'd render a float precision image so you can get all the
> out-of-focus hilight shapes just as bright as the in-focus
> hilights."
>
> I have no idea what this means or how I do it. Any ideas?
The problem is, if you apply focal blur as a post process step (as you did),
it cannot take into account any surfaces that are brighter than white in the
image file. This will make a difference around bright highlights etc. For
example, suppose that the focal blur algorithm determines that 5% of an
output pixel should be from the white surface, and 95% from the (black)
background. Using the standard image formats, this pixel would end up at 5%
grey. If you had used floating point precision, then it would be able to
take into account that the white is actually far brighter than white, and
the resulting pixel would be brighter.
Unfortunately POV does not output float precision images. It's easy to
patch it to do this, but then it's likely not going to be compatible with
your post processing.
One idea is to render the scene at (say) 10% brightness, then do your focal
blur algorithm on this image, then make it 10x brighter. You could then
merge just the highlights on top of the original or something.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
the_ajj wrote:
> Hi,
>
> I got this comment regarding one of my images 'The Chess Game' which can be
> seen in the images newsgroup:
>
> "so, did you render an 8bit image out of povray?
> Ideally you'd render a float precision image so you can get all the
> out-of-focus hilight shapes just as bright as the in-focus hilights."
>
> I have no idea what this means or how I do it. Any ideas?
If you have no idea what this means it's a bit premature to ask for a
certain solution (i.e. floating point format images).
But to shorten things: Internal focal blur works on unclipped colors,
you can generate unclipped high dynamic range output with POV-Ray (which
however does not guarantee good results with post processed focal blur)
and you can do reasonable post processed focal blur of overexposed
areas with low dynamic range images by using exponentical tone mapping
and adapting the blur function accordingly.
Christoph
--
POV-Ray tutorials, include files, Landscape of the week:
http://www.imagico.de/ (Last updated 31 Oct. 2005)
MegaPOV with mechanics simulation: http://megapov.inetart.net/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Thanks everyone!
Adrian
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
scott <spa### [at] spamcom> wrote:
> The problem is, if you apply focal blur as a post process step (as you did),
> it cannot take into account any surfaces that are brighter than white in the
> image file.
Other problems with post-processing focal blur is that it doesn't work
correctly for reflections nor refractions.
Also, true focal blur can add info to the image which is not seen in
a non-blurred image (eg. hidden surfaces can become partially visible).
Blurring a latter type of image afterwards will not add this info, naturally.
--
- Warp
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"Warp" <war### [at] tagpovrayorg> wrote in message
news:437518f2@news.povray.org
> scott <spa### [at] spamcom> wrote:
>> The problem is, if you apply focal blur as a post process step (as
>> you did), it cannot take into account any surfaces that are
>> brighter than white in the image file.
>
> Other problems with post-processing focal blur is that it doesn't
> work correctly for reflections nor refractions.
> Also, true focal blur can add info to the image which is not seen
> in
> a non-blurred image (eg. hidden surfaces can become partially
> visible). Blurring a latter type of image afterwards will not add
> this info, naturally.
Of course, but the in-built focal blur in POV isn't perfect either (no
aperture shape, finite number of samples). Which looks the "better" method
will depend on the scene. Perhaps the shape of the aperture could be added
to POV at some point.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |