CloudCompare forum

Dear community,

First of all, I would like to thank all the developers for this great open source software. You are doing a great job!

Now to my question:
The fit function "2D polygon" is an useful tool to create a surface for plane point clouds and to export the mesh file.
It really reminds me of the alphashape algorithm. Therefore, I would like to know what is behind this function "2D polygon", as I could not find a comparable function for Python. Alphashape unfortunately does not give the same results and certainly not for 3D. I would like to find a way to use a function with equal results in Python (if possible).

Thank you in advance!

HollowHeartNet
No cost too great ...

Well, it's fairly simple:
- we fit a plane on the cloud
- we project the 3D points on this plane
- we compute the 2D (concave) contour of the projected 2D points, as well as the 2D Delaunay mesh
- we back project the contour polyline and the Delaunay mesh to their '2.5D' equivalents (simply by replacing the 2D projected points by the 3D original ones)

Dear Daniel,

Thank you for the answer. Unfortunately, I'm not that talented when it comes to programming. Do you know if there is a comparable code / script for this function in Python or in another language that I can follow?

Ah, no. We only have the C++ one (but it actually uses different parts of CCCoreLib).

Does anybody know how to automate the 2D polygon fit process of a point cloud using python subprocess of CC or anything else?

The whole process I would like to automate consists of three steps:

1.) import point cloud from hard disk
2.) use 2D polygon fit function
3.) export generated mesh in OBJ format

I would appreciate any help.

daniel wrote: ↑Sat Mar 20, 2021 4:42 pm Ah, no. We only have the C++ one (but it actually uses different parts of CCCoreLib).

Can you please provide me the C++ implementation of the function "2D polygon"?
Thank you.

I guess you are referring to the "Fit facet" function?
https://github.com/CloudCompare/CloudCo ... t.cpp#L138

Yes that's what I need. Thank you, Sir!