NURBS surfaces with variable local detail level

[Yaroukh]

Hello

May I ask you guys whether something like "NURBS surfaces with variable local detail level" is within K-3D modelling ambitions?

TSplines plugin delivers this functionality in Rhino, here is how it works: http://www.youtube.com/watch?v=k1ro9S-cAwI
(The plugin can do little bit more than what is mentioned in the subject, but IMO even the %subj% by itself would be great feature to have in K-3D. To my knowledge it is not that common feature even in today's modellers, and I believe many people consider it very attractive function.)

I don't have any experience in C++, but I believe this can be done by combining two NURBS surfaces and hiding some of their geometry so that 1) the result appears as one single surface 2) user can not break the tangency at the "joining edges". (I guess K-3D would only have to keep the "joining CVs" in one line.)

What do you think about that?

Best regards
  Yaroukh

-------------------------------------------------------------------------
This SF.net email is sponsored by the 2008 JavaOne(SM) Conference
Don't miss this year's exciting event. There's still time to save $100.
Use priority code J8TL2D2.
http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone
_______________________________________________
K3d-development mailing list
K3d-development@...
https://lists.sourceforge.net/lists/listinfo/k3d-development

[hoako]

Yaroukh wrote:
> Hello
Welcome to the K-3D mailing list!

>
> May I ask you guys whether something like "NURBS surfaces with variable
> local detail level" is within K-3D modelling ambitions?
>
> TSplines plugin delivers this functionality in Rhino, here is how it
> works: http://www.youtube.com/watch?v=k1ro9S-cAwI
> (The plugin can do little bit more than what is mentioned in the
> subject, but IMO even the %subj% by itself would be great feature to
> have in K-3D. To my knowledge it is not that common feature even in
> today's modellers, and I believe many people consider it very attractive
> function.)

I don't know where I read it, but I recall reading that T-splines were
"proprietary". You may check more in tsplines doc com.
I guess this restriction wouldn't make tsplines possible on K-3D.

Although a workaround for such a tool, as you comment below, could be done.

>
> I don't have any experience in C++, but I believe this can be done by
> combining two NURBS surfaces and hiding some of their geometry so that
> 1) the result appears as one single surface 2) user can not break the
> tangency at the "joining edges". (I guess K-3D would only have to keep
> the "joining CVs" in one line.)

Cheers!
Joaquín


-------------------------------------------------------------------------
This SF.net email is sponsored by the 2008 JavaOne(SM) Conference
Don't miss this year's exciting event. There's still time to save $100.
Use priority code J8TL2D2.
http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone
_______________________________________________
K3d-development mailing list
K3d-development@...
https://lists.sourceforge.net/lists/listinfo/k3d-development

[Yaroukh]

On Fri, May 9, 2008 at 3:49 AM, Joaquín Duo <hoakoduo@...> wrote:

Yaroukh wrote:
> Hello
Welcome to the K-3D mailing list!

Hi there :o)
 

I don't know where I read it, but I recall reading that T-splines were
"proprietary". You may check more in tsplines doc com.

That is a new info for me...
http://www.tsplines.com/technology/
I don't know what can be covered by a patent. If it is possible to acquire a patent for "I have more detailed NURBS-surface here" then this thread has become a pointless one. :[

I guess this restriction wouldn't make tsplines possible on K-3D.

What exactly do you mean by "tsplines"? I hope it is not possible to own the right to provide variable-detail NURBS-surfaces - regardless of how you achieve that, just because someone gives it a single-word name...?
They are referring to some "surface intersector" and "revolutionary new" algorithms, which IMHO is far from the procedure we are talking about. I tend to think (+hope +want to believe) that T-Splines, Inc. is probably applying for a patent for something related to those push-button mesh<->tsplines translations... that is beyond the local-detail thing.

Although a workaround for such a tool, as you comment below, could be done.

It would be great!
Ofcourse only if T-Splines, Inc. won't cover the 'duplicate, trim + keep tangential' procedure with a patent. /:)

Cheers!

Joaquín 

Best regards
  Yaroukh
-------------------------------------------------------------------------
This SF.net email is sponsored by the 2008 JavaOne(SM) Conference
Don't miss this year's exciting event. There's still time to save $100.
Use priority code J8TL2D2.
http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone
_______________________________________________
K3d-development mailing list
K3d-development@...
https://lists.sourceforge.net/lists/listinfo/k3d-development

[Yaroukh]

I am sorry, I haven't noticed before that there is the bunch of PDFs: http://www.tsplines.com/technology/papers.php...
It seems now even more likely to me that their patent is something far less trivial than the mentioned workaround. :)

-------------------------------------------------------------------------
This SF.net email is sponsored by the 2008 JavaOne(SM) Conference
Don't miss this year's exciting event. There's still time to save $100.
Use priority code J8TL2D2.
http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone
_______________________________________________
K3d-development mailing list
K3d-development@...
https://lists.sourceforge.net/lists/listinfo/k3d-development

[bART Janssens-2]

On Fri, May 9, 2008 at 5:18 AM, Yaroukh <yaroukh@...> wrote:
> I am sorry, I haven't noticed before that there is the bunch of PDFs:
> http://www.tsplines.com/technology/papers.php...
> It seems now even more likely to me that their patent is something far less
> trivial than the mentioned workaround. :)

Hi Yaroukh,

>From their about page: "T-Splines, Inc. was founded in 2004. The
company holds an exclusive license to the patented T-Splines
technology developed by Dr. Thomas Sederberg at Brigham Young
University." See also
http://www.patentstorm.us/patents/7274364-claims.html

Seems the only way to use TSplines legally, is to buy a license to
their software library.

I think it's loathsome that public funding can be used to do research
for something that ends up as an exclusive technology for one company
making money off it.

Cheers,

--
Bart

-------------------------------------------------------------------------
This SF.net email is sponsored by the 2008 JavaOne(SM) Conference
Don't miss this year's exciting event. There's still time to save $100.
Use priority code J8TL2D2.
http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone
_______________________________________________
K3d-development mailing list
K3d-development@...
https://lists.sourceforge.net/lists/listinfo/k3d-development

[Yaroukh]

On Mon, May 12, 2008 at 11:59 AM, Bart Janssens <bart.janssens@...> wrote:

On Fri, May 9, 2008 at 5:18 AM, Yaroukh <yaroukh@...> wrote:
> I am sorry, I haven't noticed before that there is the bunch of PDFs:
> http://www.tsplines.com/technology/papers.php...
> It seems now even more likely to me that their patent is something far less
> trivial than the mentioned workaround. :)

Hi Yaroukh,

>From their about page: "T-Splines, Inc. was founded in 2004. The
company holds an exclusive license to the patented T-Splines
technology developed by Dr. Thomas Sederberg at Brigham Young
University." See also
http://www.patentstorm.us/patents/7274364-claims.html

Seems the only way to use TSplines legally, is to buy a license to
their software library.

...only if you actually intend to use their algorithms I guess. TSplines provide more functionality than %subj%, and I referred to that particular plugin only to explain what I mean by "variable detail level". (I don't know whether there is a name for such type of surface, also English is not my mother tongue; so I only wanted to show how it may look.) I believe the whole idea is older than the TSplines plugin ot the company.

If it is possible to "simulate" somehow the variable detail level for NURBS-surfaces the way I have suggested then you wouldn't need to care about TSplines.
(Yet I have to admin I'm not 100pct sure it is possible that way; I mean technically, regardless of any patents.)

Ofcourse if you are saying that providing variable detail level for NURBS-surfaces inevitably means violating TSplines patent than it is over ofcourse. (But I doubt that because that would be like rejecting LCD because there is already OLED.)

I think it's loathsome that public funding can be used to do research
for something that ends up as an exclusive technology for one company
making money off it.

It sounds weird to me too..

Cheers,

--
Bart

Best regards
  Yaroukh

-------------------------------------------------------------------------
This SF.net email is sponsored by the 2008 JavaOne(SM) Conference
Don't miss this year's exciting event. There's still time to save $100.
Use priority code J8TL2D2.
http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone
_______________________________________________
K3d-development mailing list
K3d-development@...
https://lists.sourceforge.net/lists/listinfo/k3d-development

[Daniel S. Matthews]

On Mon, May 12, 2008 at 8:42 PM, Yaroukh <yaroukh@...> wrote:

> On Mon, May 12, 2008 at 11:59 AM, Bart Janssens <bart.janssens@...>
> wrote:
>>
>> On Fri, May 9, 2008 at 5:18 AM, Yaroukh <yaroukh@...> wrote:
>> > I am sorry, I haven't noticed before that there is the bunch of PDFs:
>> > http://www.tsplines.com/technology/papers.php...
>> > It seems now even more likely to me that their patent is something far
>> > less
>> > trivial than the mentioned workaround. :)
>>
>> Hi Yaroukh,
>>
>> >From their about page: "T-Splines, Inc. was founded in 2004. The
>> company holds an exclusive license to the patented T-Splines
>> technology developed by Dr. Thomas Sederberg at Brigham Young
>> University." See also
>> http://www.patentstorm.us/patents/7274364-claims.html
>>
>> Seems the only way to use TSplines legally, is to buy a license to
>> their software library.
>
> ...only if you actually intend to use their algorithms I guess. TSplines
> provide more functionality than %subj%, and I referred to that particular
> plugin only to explain what I mean by "variable detail level". (I don't know
> whether there is a name for such type of surface, also English is not my
> mother tongue; so I only wanted to show how it may look.) I believe the
> whole idea is older than the TSplines plugin ot the company.
>
> If it is possible to "simulate" somehow the variable detail level for
> NURBS-surfaces the way I have suggested then you wouldn't need to care about
> TSplines.
> (Yet I have to admin I'm not 100pct sure it is possible that way; I mean
> technically, regardless of any patents.)
>
> Ofcourse if you are saying that providing variable detail level for
> NURBS-surfaces inevitably means violating TSplines patent than it is over
> ofcourse. (But I doubt that because that would be like rejecting LCD because
> there is already OLED.)
>
>> I think it's loathsome that public funding can be used to do research
>> for something that ends up as an exclusive technology for one company
>> making money off it.
>
> It sounds weird to me too..
>
>> Cheers,
>>
>> --
>> Bart
>
>
> Best regards
>   Yaroukh
>

What you need is a spline surface subdivision tool. Take a patch and
split it into multiple seamless patches while retaining a "memory" of
the mother patch so that when a child patch is manipulated you can
calculate the required adjustments to the family of patches so that
they remain seamless. Doing this in a hierarchical manner would allow
any level of detail. Managing texture mapping over the entire family
tree of the spline patches would be required too.

Is that in essence what you require? If I have just reinvented the
above mentioned patent then I would not worry to much about the patent
as it is so obvious. If what I have described is not covered by the
patent then you just need to do the math to work out how to write the
code library that such a tool set would be built upon. Sorry I can't
help you with the applied math, but I hope the theory will point you
in the right direction.

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
K3d-development mailing list
K3d-development@...
https://lists.sourceforge.net/lists/listinfo/k3d-development

[Yaroukh]

What you need is a spline surface subdivision tool. Take a patch and
split it into multiple seamless patches while retaining a "memory" of
the mother patch so that when a child patch is manipulated you can
calculate the required adjustments to the family of patches so that
they remain seamless. Doing this in a hierarchical manner would allow
any level of detail. Managing texture mapping over the entire family
tree of the spline patches would be required too.

Is that in essence what you require?

Yes, and that is actually what I wrote in the very first message of this thread. :) I only haven't had enough time to test it by playing with good old NURBS-surfaces. (To make sure that having joining CVs in one line is the only thing required to keep the edges joined seamlessly at all conditions.)
 

If I have just reinvented the above mentioned patent then I would not worry to much about the patent
as it is so obvious.

I agree and hope so!

Best regards
  Yaroukh

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
K3d-development mailing list
K3d-development@...
https://lists.sourceforge.net/lists/listinfo/k3d-development