Can I generate a Point Cloud from a grid?

Question

I am trying to create point cloud data from a grid, such as a (.obj) Maya file. But I could only find the opposite case, a point cloud in a grid on the Internet. Is there a way to create “cloud point data from a grid” using 3D tools like MeshLab or Maya? (I prefer to use MeshLab)

Thanks.:)

Answer 1

Cloud Cloud Library has several different command line tools for turning meshes into point clouds, as far as I know, an object into points from a set of views and combining visualizations. e.g. pcl_mesh2pcd, pcl_mesh_sampling

Answer 2

Interestingly, there are no definitive answers to this question on the Internet.

Converting from a grid to a point cloud is not like clogging the vertices of a grid into a point cloud! Mesh is a rare representation of a point cloud. Therefore, in order to convert the grid to a point cloud, you need to try the points on the surface of the grid. PCL has a utility for execution called pcl_mesh_sampling . Source code is located here .

In principle, samples of methods N uniformly displayed from the surface of the grid using VTK. It is very effective and you will get point clouds with any number of points that you want.

Answer 3

Converting a triangulated mesh to a point cloud means that you want to make an exact selection on that surface. Depending on the required distribution (for example, Uniform Montecarlo, Poisson Disk, etc.), there are different algorithms with completely different results. For example, if you need a well-distributed random point, you need a Poisson disk distribution. You can test some of these algorithms inside meshlab or directly on your browswer at http://www.meshlabjs.net/ (just load the grid and enter “fetch” in the search bar).

You can find the open source C ++ implementation inside the vcg mesh processing library ( http://www.vcglib.net ) and a description of this algorithm (quite simple to implement):

Efficient and flexible sampling with blue noise properties triangular mesh
M. Corsini, P. Signigni, R. Skopinho
IEEE transactions for visualization and computer graphics 18 (6), 914-924
http://vcg.isti.cnr.it/Publications/2012/CCS12/

Answer 4

CloudCompare, a free open source project , can load OBJ grids and then generate cloud points on the grid, which can then be saved in various formats.

Answer 5

I was looking for a similar problem. I needed to design a cell model in different poses to generate pointcloud data.

And then I found below the project on github. If someone wants to create pointclouds (.pcd) from a mesh object (.obj) in different poses, this will help.

This will simulate a pointnoud of kinect format for a given mesh object.

https://github.com/jbohg/render_kinect

Answer 6

I wanted to do the same, but preferred something that could be easily done in Python. I found this post to be really helpful. It uses a Python library called pyntcloud . It loads the grid in .ply format and selectively displays a certain number of points from it, so that it covers the surface in a more or less uniform way. It can be easily adapted for other formats. I did this for .off files and it works well.