3
$\begingroup$

I'd like to know if there's any consensus on naming mesh data as part of the mesh object.

For example, if one takes a monkey object (suzanne), it consists of 3 chunks of linked geometry - 1 monkey head and 2 eyes.
Is it called that monkey object is composed of 3 meshes or is it one mesh and if one would like to count how many parts there are it would require to refer to those parts as chunks of linked geometry?

There is article in Wikipedia but it doesn't mention whether mesh is chunk of linked geometry or not:

A polygon mesh is a collection of vertices, edges and faces that defines the shape of a polyhedral object in 3D computer graphics and solid modeling
https://en.wikipedia.org/wiki/Polygon_mesh

There is also related information in this gamedev post but it doesn't specifically mention if mesh is only connected chunk of geometry (vertices, edges, faces) or all geometry of one mesh object.

So in essence, does "mesh" word mean only chunk of linked geometry as part of mesh object or all geometry defining that mesh object? Is this naming convention specific to Blender or not?

$\endgroup$
11
  • 2
    $\begingroup$ Suzanne can be considered one mesh that consists of multiple disconnected islands. It's therefore a non-manifold mesh. Whether disconnected islands are considered to be part of the same mesh, depends on how the data is currently stored and treated by the software. The semantics aren't specific to Blender, but computer graphics in general. Posting this as a comment, because an answer would need to go into more depth. I can post a full answer if this solves your problem. $\endgroup$
    – Robert Gützkow
    Mar 2, 2020 at 15:05
  • $\begingroup$ Personally I lean towards the latter. $\endgroup$
    – batFINGER
    Mar 2, 2020 at 15:37
  • 1
    $\begingroup$ It's not just a convention, for Blender? It's a useful functional definition. Meshes have internally uniquely indexed vertices, which can share edges, they all have locations defined in their object's space, carry their object's transforms, etc. $\endgroup$
    – Robin Betts
    Mar 2, 2020 at 18:53
  • 1
    $\begingroup$ I didn't expect this to be so difficult to answer. The word "mesh" is hardly ever formalized in the literature and when it is, it's usually in a specific context, e.g. a two manifold triangle mesh. I'm afraid I can't give you a definitive answer, since apparently there isn't one. What I can definitely say is, that the separated faces are commonly called "island" or "isle". Whether or not they are considered to be a part of the same mesh seems to depend on your own definition of what a mesh is and the data structure you choose to represent it. The same applies to singular vertices. $\endgroup$
    – Robert Gützkow
    Mar 4, 2020 at 20:55
  • 1
    $\begingroup$ Geometric Modeling Based on Polygonal Meshes may give you some answers to your question. For the wide variety of meshes than can be considered, see the questions in the section Topological requirements (p. 17). For an overview of artifacts in non-manifold meshes see figure 4.1 (p. 26). For the definition of manifold/non-manifold see p. 9. $\endgroup$
    – Robert Gützkow
    Mar 4, 2020 at 21:03

0

You must log in to answer this question.

Browse other questions tagged .