# comp.graphics.algorithms

## Subject: Re: cell complex indexing

Thanks for the response. I believe you misunderstood my question. I need
to store data on each cell of the complex, not just the voxels (i.e. data
on vertices, edges, faces, and voxels).

What I ended up doing was to subdivide the grid once (octree style) so I
have a vertex for each cell of the original grid. Index this subdivided
grid normally. Then the dimension of a cell is determined by the number of
odd indices, and you can get the facets/cofacets by adding or subtracting
integer unit vectors depending on which dimensions are odd/even.

Though this is a simple solution, some colleagues suggested augmenting
the (i,j,k) index with an additional facet index which would be
cumbersome.

best
- Taylor

On Wed, 26 Apr 2006, Hans-Bernhard Broeker wrote:

> W. Taylor Holliday wrote:
> > Hi Folks,
>
> > Question: what is a good way of indexing a rectilinear voxel grid (3d
> > cell complex) such that you have an index for each cell (vertex, edge,
> > quad face, cube)? and can easily find the facets and co-facets (incidence
> > relationships) of a cell?
>
> The obvious one: cartesian coordinates, e.g. a 3D array of voxels. If
> you can't handle 3D arrayas, you can always map the triplet of indices
> to a single index:
>
> [i][j][k] --> [J*K*i + K*j + k]
>
> --
> Hans-Bernhard Broeker (broeker@physik.rwth-aachen.de)
> Even if all the snow were burnt, ashes would remain.
>

___________________________________________________________
W. Taylor Holliday wtholliday.org