refine-mesh v1.0.1

refine-mesh

Iterative mesh refinement based on the technique in

• M. Botsch, L. Kobbelt. 2004 "A Remeshing Approach to Multiresolution Modeling" EUROGRAPHICS 2004

Give this some garbage mesh as input and it will automagically give you a nice uniformly subdivided mesh.

Example

var refine = require('refine-mesh')
var bunny = require('bunny')

var normals = require('normals').vertexNormals(bunny.cells, bunny.positions)

//Refine the bunny from a base mesh to one with a uniform edge length of 0.1
var refined = refine(
  bunny.cells,
  bunny.positions,
  normals,
  {
    edgeLength: 0.1
  })

So, a coarse bunny mesh like this:

Becomes a fine mesh like this:

Install

npm i refine-mesh

API

var mesh = require('refine-mesh')(cells, positions, normals[,options])

Applies iterative mesh refinement to a cell complex

• cells is an array of triples representing the faces of the mesh
• positions is an array of [x,y,z] triples representing the vertices
• normals is an array of [nx,ny,nz] triples representing the normals
• options is an object with the following properties:
  • edgeLength is the desired edge length (default is median edge length)
  • maxIters is the maximum number of iterations (default 5)
  • splitIters maximum number of edge split iterations (default 10)
  • collapseIters maximum number of edge collapse iterations (default 10)
  • flipIters maximum number of edge flip iterations (default 10)
  • smoothIters maximum number of smoothing iterations (default 20)
  • smoothRate constant controlling rate of smoothing (default 0.95)
  • minSplit minimum number of splits per iteration (default 0)
  • minCollapse minimum number of collapse operations (default 0)
  • minFlip minimum number of flip operations (default 0)
  • minSmooth minimum smoothing improvement (default 0.01*edgeLength)

Returns An object with the following properties

• cells the faces of the resulting mesh
• positions vertex positions of result
• normals vertex normals of result

License

(c) 2015 Mikola Lysenko. MIT License