Splats a point cloud onto a volumetric grid.
var splat = require("splat-points-3d")
var pack = require("ndarray-pack")
var zeros = require("zeros")
var points = pack([
[1, 0, 0],
[0, 4, 0],
[0, 0, 8]
])
var weights = pack([
1,
2,
4
])
var grid = zeros([32, 32, 32])
splat(grid, points, weights, 3.0)
npm install splat-points-3d
require("splat-points-3d")(out, points, weights, radius)Splats a collection of points onto a 3D volumetric grid using Lanczos/sinc filtering. If you want a different kernel afterwards, you can apply a convolution via the FFT
out is a 3D ndarraypoints is a [n, 3] shape ndarray encoding the x/y/z grid coordinates of the points to splatweights is a [n] shape ndarrayradius is the radius of the point splat to draw(c) 2014 Mikola Lysenko. MIT License