Firstly, I have managed to fix some instances of missing polygons in the Delaunay triangulation which were resulting from QHull not triangulating the simplexes properly.
I have implemented my idea for removing interior polygons but have found that triangles added by the delaunay triangulation are not attaching to existing edges which comes as quite a surprise and is the final nail in the coffin for the Alpha Hulls/Shapes methods unless a revival can be performed tomorrow when talking with Tyrone.
In preparation for moving on I have followed up on the earlier suggestion of progressive meshes, reading papers about view-dependent and view-independent methods as well as the skip strip optimisations to improve performance of mesh changes at run time. While I do believe I understand the processes behind them, I'm not 100% confidant that I have enough time remaining to implement either of them.
I have modified the viewing application to give the user more control over the environment but I have yet to populate the scene sufficently.
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment