coplanar data input -> corner effect

Feb 27, 2014 at 5:38 PM
Hi,
i follow up to the Input vertex degeneracy post some time ago.
maybe you can give me a advice to approach this.
Your adaption to the last push solved a lot but the corners still not always get considered.

Image

Adding noise to the input data solves the connectivity but of course changes the border tetrahedrons in the convexhull.

Image

do you have any idea to resolve the corners, to receive all delaunay cells?
Feb 27, 2014 at 5:43 PM
below the input data {x,y,z}
input data
Developer
Feb 28, 2014 at 12:50 AM
Hey,

if you know the data is a lattice, maybe it would not be that hard to write your own triangulation routine? A cuboid splits into 6 tetrahedrons, it should not be that hard to write a routine that tiles so that the adjacency fits.

Anyway, if I get some time, I will have a look if I can find some trick to solve the boundary points -- tho no promises.
Feb 28, 2014 at 8:44 AM
Hey,
Thanks for your reply. Indeed it wouldn't be so hard to write the triangulation routine for this standard case. I used just this one to show the problem i have in some cases when they reach coplanarity in the corner. Would be great if you find some time. If you need anything let me know.