Introduction to ‘Automatic Polygonal Mesh Healing’

A little intro into the project. BRL-CAD’s manifold meshes are represented by the type called bag of triangles (bot). The meshes are not perfect always, they might become defective. It might happen during import-export or during *, etc. And defective meshes are not desirable. So, we have a a problem here. And the solution is my project.

Specifically we focus on the geometric defects such as gaps, overlaps, T-joints, and holes, and topological defects such as singular vertices, singular edges, and inconsistent orientation of the triangles in the mesh.

A little while after i had submitted the proposal, on the BRL-CAD IRC channel, OpenSCAD’s Marius Kintel (nick: kintel) asked if there could be a portable module for mesh healing that could be used by both BRL-CAD and OpenSCAD. After my proposal got accepted, we decided to go forward with the portable module idea.

Now, how we’d implement the portable module was of interest to all of us. Sean and Daniel from BRL-CAD suggested a few methods and we discussed the pro and cons of both.

At this point my mentor Daniel had suggested we go with a generic container of an intermediate polygonal mesh using non-generic programming. That is we define an intermediate structure that will be used to complete the mesh healing functionalities. And we implement the structure differently for the two organizations. We do them using a set of functions that get the attributes from the native structures. In this way, we have an intermediate structure that can talk to both the native structures.

The method i had suggested, in essence, was to create a structure that is in a sense a “union” of the attributes of both the native structures. The only conversion process here will be to set those attributes of the intermediate structure to those of each organisation (the ones that apply). For the attributes that don’t apply to one of the organisations, we set it to NULL. In this way too, we have an intermediate mesh that encompasses the properties of both the meshes and one that can be used in the portable module.

At this point, we are yet to decide on the method to go forward with.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s