 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
I'm a bit fuzzy on how to transform a list of coordinates into a mesh. What
would be the easiest way to generate a mesh based on a table of 3D vectors?
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
How Camp wrote:
> I'm a bit fuzzy on how to transform a list of coordinates into a mesh. What
> would be the easiest way to generate a mesh based on a table of 3D vectors?
>
There is no simple solution, because you've lost the connectivity information.
Best approach is probably to use the 3D coordinates to generate a scalar
field, and then to generate the mesh out of the scalar field.
see http://astronomy.swin.edu.au/~pbourke/modelling/polygonise/
Beware, if you are relevant to USA patent, you should check the marching
cube and other patents before going on.
--
Non Sine Numine
http://grimbert.cjb.net/
Puis, s'il advient d'un peu triompher, par hasard,
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Thanks, L'Harmonieux, for the information. I've seen various 3D contour
sofware packages perform some operations on a table of XYZ coordinates, and
produce interpolated mesh results. For example, SigmaPlot, Surfer for
Windows, etc. However, they provide no explanation as to their methods.
What if I wanted to 'fake' the connectivity info? For example, choosing
between linearly connecting all the data points, or some non-linear fit.
Would it still be best to go the route of creating a scalar field?
- How
"L'Harmonieux Forgeron" <jgr### [at] free fr> wrote in message
news:3CE### [at] freefr...
> How Camp wrote:
>
> > I'm a bit fuzzy on how to transform a list of coordinates into a mesh.
What
> > would be the easiest way to generate a mesh based on a table of 3D
vectors?
> >
> There is no simple solution, because you've lost the connectivity
information.
>
> Best approach is probably to use the 3D coordinates to generate a scalar
> field, and then to generate the mesh out of the scalar field.
>
> see http://astronomy.swin.edu.au/~pbourke/modelling/polygonise/
>
> Beware, if you are relevant to USA patent, you should check the marching
> cube and other patents before going on.
>
> --
> Non Sine Numine
> http://grimbert.cjb.net/
> Puis, s'il advient d'un peu triompher, par hasard,
>
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> What if I wanted to 'fake' the connectivity info? For example, choosing
> between linearly connecting all the data points, or some non-linear fit.
> Would it still be best to go the route of creating a scalar field?
I've tried this, but didn't have the time to finish. The idea behind my
program was simple to build a small pyramid and then enlarge this pyramid,
adding the best fitting points. But my program lacks some tests and so there
are still some possibilities (for example it doesn't check if the connecting
pyramid hits the existing body). I think with some work it could be usefull,
but my source code (written in C) is very messy and not documented (or only
in German).
But if you want to have a look at it, you're welcome.
Best regards,
Florian
Oh, btw I have posted this problem to povray.newusers some time ago and got
some useful replies, have a look at that thread...
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
|
|
