 | | | UV pelt, Unwrap | UV pelt, Unwrap 2004-03-05 - By Graham D Clark
Back
But that's one of the reasons for pelting it, as it preserves the
individual poly proportions as much as possible with little stretch of
the UVs and importantly the ability arrange the UVs via pelt method as
its easier and more logical and natural to the form, to texture it.
We have an external tool to do it now and of course there is Houdinis
built in one that is good and fast but is missing some things that
require painting weight/density attribute that instead being predictable
for legs and we'd like to just automate.
We have some issue with poly/point ordering soemwhere in the pipeline so
I was looking for the basic concepts behind how UV pelting could work
and in interest to make my own in XSI to also add some things that might
be useful.
I'm a little sidetracked now anyway to script match UVs on disimilar
ordered meshes :/ but will come back to this soon as I think it could be
quite useful and maybe faked with OPs to fake the heavy math.
-- ---- ---- ---- ---- ---- --
http://www.grahamdclark.com
> -- --Original Message-- --
> From: owner-xsi@(protected)
> [mailto:owner-xsi@(protected)] On Behalf Of kim aldis
> Sent: March 5, 2004 11:36 AM
> To: XSI@(protected)
> Subject: RE: [script and Dev] UV pelt, Unwrap
>
>
> You shouldn't expect too much from this, not because it's a
> bad idea but
> because uvs don't take too kindly to being stretched when
> rendered using
> mental ray. For an example, try this:-
>
> Get a one poly grid, put a texture projection on it in x/z
> and a noicon
> texture.
> Now, go to the texture editor, select all uvs and scale down
> to s0-30%. Now
> take one corner point and pull it back out somewhere near
> it's original
> position and preview the grid.
>
> > -- --Original Message-- --
> > From: owner-xsi@(protected)
> > [mailto:owner-xsi@(protected)] On Behalf Of Graham D Clark
> > Sent: 04 March 2004 13:37
> > To: XSI@(protected)
> > Subject: RE: [script and Dev] UV pelt, Unwrap
> >
> > Thanks to all for the links and key words to search for.
> > I guess due to the shared connection we have here I cannot
> > get to that pdf link in previous message below without an
> > error, but using the terms you guys have given me will be
> > able to find some good leads.
> > I think my first basic test approach to this will be
> > rebuilding the UV set as a mesh then using operators on it
> > for interactive adjustment, as I already have the simple UV
> > to mesh part and a scop to send it back to UV set.
> > Thanks again for the leads,
> > Graham
> > -- ---- ---- ---- ---- ---- --
> > http://www.grahamdclark.com
> >
> >
> >
> > > -- --Original Message-- --
> > > From: owner-xsi@(protected)
> > > [mailto:owner-xsi@(protected)] On Behalf Of Damon Godley (Core
> > > Design Ltd.)
> > > Sent: March 4, 2004 10:36 AM
> > > To: XSI@(protected)
> > > Subject: RE: [script and Dev] UV pelt, Unwrap
> > >
> > >
> > > Hi,
> > >
> > > This article refers to a version done in maya, which
> > although tedious
> > > can be done by hand using dynamics. I wrote a mel script
> > that set it
> > > up, and I know of a few other people that where playing with this
> > > idea. That said it is relatively easy to miss out the
> dynamics and
> > > make your own solver and set it up to be more specific to
> what you
> > > need to achieve, and as I can see would be easy to
> > implement into XSI.
> > > If you do some more searching though you will find that
> > there are more
> > > advanced method's of pelting.
> > >
> > > Try searching for UV parameterization, least squares
> > conformal mapping
> > > , Automatic Texture Atlas Generation etc.
> > >
> > > This all gets a bit hard though and my interest was
> dazzled by the
> > > complicated math involved.
> > >
> > > I still kept going with the spring method and have some
> reasonable
> > > results.
> > > If you want any more help on that method give me a shout,
> though if
> > > you can handle more advance subjects search for the above
> > as they will
> > > in long run provide more satisfying results.
> > >
> > > Hope that helps
> > >
> > > Damon.
> > >
> > > -- --Original Message-- --
> > > From: owner-xsi@(protected)
> > > [mailto:owner-xsi@(protected)]On Behalf Of Ben Kilgore
> > >
> > > Sent: 04 March 2004 08:58
> > > To: XSI@(protected)
> > > Subject: Re: [script and Dev] UV pelt, Unwrap
> > >
> > >
> > > The incomparable Borshukov @ Siggraph 2000:
> > >
> > http://www.cs.berkeley.edu/~gdb/publications/acrobat/pelt.pdf.gz
> > -- -- Original Message -- --
> > From: "Graham D Clark" <mail@(protected)>
> > To: <xsi@(protected)>
> > Sent: Wednesday, March 03, 2004 2:46 PM
> > Subject: [script and Dev] UV pelt, Unwrap
> >
> >
> > > Does anyone have any links for the concept of writing a UV
> > unwrapping
> > > via pelt method? Perhaps with mention of tension, spring edge and
> > > reference components? I'm not finding too much online in
> > early online
> > > search, perhaps any technical document on it might also
> give me some
> > > more of the key terms to search for and find some that
> > actually cover
> > > the technical aspects instead of the manual approaches.
> > > Any ideas appreciated.
> > > Thanks, Graham
> > > -- ---- ---- ---- ---- ---- --
> > > http://www.grahamdclark.com
> > >
> > >
> > >
> > ---
> > Unsubscribe? Mail Majordomo@(protected) with the following text in
> > body:
> > unsubscribe xsi
> >
> > ---
> > Unsubscribe? Mail Majordomo@(protected) with the following text in
> > body:
> > unsubscribe xsi
> >
> >
> >
> >
> > ---
> > Unsubscribe? Mail Majordomo@(protected) with the following
> > text in body:
> > unsubscribe xsi
> >
> >
> >
>
>
>
> ---
> Unsubscribe? Mail Majordomo@(protected) with the following
> text in body:
> unsubscribe xsi
>
>
>
---
Unsubscribe? Mail Majordomo@(protected) with the following text in body:
unsubscribe xsi
|
|
|