swiftcoder/isosurface

Rust algorithms for isosurface extraction

isosurface's Issues

Rename the torus example

The torus example has morphed into a more complete sample application, as it now encompasses multiple algorithms and distance fields. It would be nice to find it a name that better reflects its current identity.

Would you be open to additional code commenting?

I have been using this repo as a guide while learning about marching cubes / oct tree duals / Morton codes. I appreciate the elegance in your implementation, in particular the use of closures to promote code reuse. I have also been reading through the paper you linked, as well as some of their sources, as I'm not sure I could have connected everything on my own.

As I learn, I have been adding comments to the code to help me solidify my understanding. Given that this the only rust library of its kind (that I can find) I thought it might be useful to others to have these notes (particularly if they don't have access to the academic articles in question.) Would you be open to a pull request adding my additional code comments? I would of course be willing to edit them to your satisfaction. I understand either way!

Marching cubes index caching may be broken

The linear hashed marching cubes algorithm is producing meshes with roughly the same number of triangles, but half the number of unique vertices compared to the basic marching cubes algorithm.

This likely means that my (overcomplicated) index cache isn't actually deduplicating all of the vertices that it should be. This issue is to verify whether the index cache is actually working correctly, and fix it if necessary. If it makes sense to simplify the index cache logic in the process, that'd be a general win.