NURBS with extraordinary points: High-degree, non-uniform, rational subdivision schemes

[chucktesta1997]

yeeeeee nurbs

[DrSchille]

Very interesting. I've been looking for something like this. Well done!

[KRAZY7777]

what are b-splines :P

[RayTomlin]

Can this be implemented for Blender? Or is there a Linux platform that this can be run on?

[CashmanTom]

@RayTomlin The source code (available from the link in the video description) can be compiled for Linux platforms. The open source modeller 'Ayam', which runs on Linux, has also incorporated our example code as an experimental plugin. Of course this method _could_ also be implemented for Blender, but that depends on the priorities for the Blender development team!

[StephenTCooney]

Great work.

[alainvig]

what does "C1 continuity" means ?

[CashmanTom]

@alainvig It's a measure of parametric continuity which indicates that the first derivative is continuous. More information is available by searching for "C1 continuity" on Google, for example.

[alainvig]

@CashmanTom okay...do you have the source code for this ?

[CashmanTom]

@alainvig Implementation, source code and preprint for this paper are all available from the link in the video description.

[gururedbull]

wonder how this is different from TSplines?

[CashmanTom]

@gururedbull There are several differences. The scheme described in our paper cannot include "T-junctions" in the mesh: the key benefit of T-Splines. But T-Splines are unable (as far as I am aware) to handle high polynomial degrees in the way we do here. Both representations are compatible with NURBS, in some sense, but I think T-Splines are a superset of only bicubic NURBS, while our paper gives a superset of NURBS without any restriction on degree or knot spacing.

[4jeremy9]

beautiful