[Bf-cycles] Interactive GI filtering paper / Siggraph 2014

[Brecht Van Lommel]

I think bidirectional methods will at some point become the standard for
production rendering. It's certainly something that would be great to have
in Cycles.

There are various unsolved problems there still, handling outdoor scenes,
image caches,  how to apply certain tricks in this context, etc.

It might take a while still before this gets used for animation a lot,
because you're not so much speeding up rendering, but rather making
difficult light setups possible that would now be avoided.

Anyway, very interesting stuff to keep an eye on.

Brecht.

On May 10, 2014 8:30 PM, "Marco G" <digitalbath86 at hotmail.com> wrote:
>
> Ah well i thought this one was a lot better than other similar or offline
methods.
> Regarding such approximations (irradiance caching) i tried to involve a
couple of guys in the past months but without success yet.
>
> Approximations aside, the upcoming Siggraph seems will have a nice focus
on
> more advanced light transport, focusing on usability from the VFX and
Architecture side too,
> and indeed such approximation seems are going to disappear in some way,
> or at least progressively abandoned...according to these slides path
tracing is the most used
> but it seems industry proven renderers are considering/investigating more
recent and advanced methods too. (see next link) **
> In part it's a matter of "breaking" artists workflows again, and the good
visual feedback unidirectional PT provides.
>
> Perhaps simple path tracing will be replaced by some BiDir derivate, how
you see this trend?
> There will be a course held by Pixar, Fajardo, the Corona render guy,
Nvidia, Next Limit...
> ** http://www.iliyan.com/publications/Siggraph2014Course
> ** Syllabus
>
> One of the guy behind Vertex Connection and Merging is currently working
at Solid Angle,
> and VCM is going to hit Renderman 19 too (and already integrated in Vray
and Corona, see link, last sildes)
>
http://cgg.mff.cuni.cz/~jaroslav/papers/2014-ltstutorial/04%20-%20ltstutorial%20-%20ppm+bpt%20-%20georgiev%20-%20notes.pdf
>
> By the way, a first implementation of VCM on GPU has appeared,
> and this link in particular also focus on making difficult light
transport algorithms more efficient on GPUs.
> http://cgg.mff.cuni.cz/~jaroslav/papers/2014-gpult/index.htm
>
> Lastly, found this one which improve BiDir dramatically according to the
examples, but still unreleased.
> http://cgg.mff.cuni.cz/~jaroslav/papers/2014-onlineis/index.htm
>
> So to conclude, BiDir or some of its evolution will be the next
state-of-the-art approach?
> How feasible will be for Cycles to switch to BiDir in the future, if it
will be proven to be the best choice? (given enough man-power)
>
> Regards,
> MG
>
> > From: brechtvanlommel at pandora.be
> > Date: Wed, 7 May 2014 00:03:28 +0200
> > To: bf-cycles at blender.org
> > Subject: Re: [Bf-cycles] Interactive GI filtering paper
> >
> > There's a lot of such filtering papers and it's closely related to
> > irradiance caching. Certainly such methods would be useful for faster
> > previews and rendering still images. There's simple algorithms and
> > complex ones with different tradeoffs and inaccuracies. If people want
> > to contribute code for this kind of thing that can fit fine.
> >
> > For animations however any method that does such filtering is tricky
> > to get working without flickering / low frequency noise. The new
> > papers in this area do not actually improve accuracy / reliability
> > over irradiance caching as far as I can tell, they're worse but more
> > general or quicker. Irradiance caching with all the tricks that have
> > been developed for it over the years might still be the least flickery
> > of all for the problem of diffuse indirect light.
> >
> > I don't know of other reliable algorithms for animation that also
> > provide a big speedup for the simple cases. Mostly it seems that
> > Cycles could handle the difficult geometry, shader lighting cases
> > better with various changes, but if you want a big speedup for the
> > simpler cases then some sort of filtering or caching method with the
> > associated tradeoffs is the only solution I know of.
> >
> > Brecht.
> >
> >
> >
> >
> > On Tue, May 6, 2014 at 10:35 PM, Marco G <digitalbath86 at hotmail.com>
wrote:
> > > Hi all,
> > > just stumbled upon this paper, i'd be curious to know what Cycles
devs think
> > > about such technique,
> > > results seems quite accurate and unlike other filtering/reconstruction
> > > methods, it is done online interactively.
> > >
> > > link -
http://graphics.berkeley.edu/papers/Udaymehta-IPB-2013-07/index.html
> > >
> > > Would something like this be unusable, introducing a lot of
inaccuracy? (or
> > > complexity to the code..?)
> > > Speaking more in general, isn't there any idea (for the future) to
> > > help/simplify diffuse-indirect calculation? (not Irradiance cache
related,
> > > but something animation-friendly)
> > >
> > > Thanks,
> > > regards.
> > >
> > > _______________________________________________
> > > Bf-cycles mailing list
> > > Bf-cycles at blender.org
> > > http://lists.blender.org/mailman/listinfo/bf-cycles
> > >
> > _______________________________________________
> > Bf-cycles mailing list
> > Bf-cycles at blender.org
> > http://lists.blender.org/mailman/listinfo/bf-cycles
>
> _______________________________________________
> Bf-cycles mailing list
> Bf-cycles at blender.org
> http://lists.blender.org/mailman/listinfo/bf-cycles
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.blender.org/pipermail/bf-cycles/attachments/20140510/63b0b057/attachment.htm