[vtk-developers] howto report a bug?

[Henning Meyer]

I'm actually thinking of skipping sampling altogether and move to
direct polygonization using approaches like:
http://www.sciencedirect.com/science/article/pii/S0097849305001317

Something like this would also make a nice addition to VTK, wouldn't it?
Unfortunately the referred algorithms need Hessians of the implicit
function and deriving the Hessian symbolically is currently bending my
brain...

2012/10/6 Will Schroeder <will.schroeder at kitware.com>:
> This is a very cool and ambitious project. Chances are you are going to
> rapidly run into resolution issues (if for example you sample the functions
> on volumes) due to the uniform partitioning of space of volumetric sampling.
> You may have to use an adaptive sampling process, say driven by something
> like an octree. And hopefully you've had a chance to review some of the
> literature out there like
> http://www-evasion.imag.fr/Membres/Franck.Hetroy/Teaching/ComputerGraphics/Articles/bloomenthal_wyvill-i3d1990.pdf
> (a good starting point).
>
> W
>
>
> On Sat, Oct 6, 2012 at 12:59 AM, Henning Meyer <tutmann at gmail.com> wrote:
>>
>> > As a work around, just scale the plane uwing SetNormal so that the range
>> > of the data is the same as the sphere (or other data). No need to implement
>> > your own.
>>
>> That would work for the synthetic examples I made up. But I need a general
>> solution as I'm developing a general purpose CAD based on Implicit
>> Functions.
>
>
>
>
> --
> William J. Schroeder, PhD
> Kitware, Inc.
> 28 Corporate Drive
> Clifton Park, NY 12065
> will.schroeder at kitware.com
> http://www.kitware.com
> (518) 881-4902