-1

I am working on the preparation of training data for point cloud segmentation.

I generated nearly 4000 point cloud patches by cropping them with footprints.

Since the size of footprints is variable, my patches have different amounts of points. Some of them contain 10.000 points, and some 1000.

My goal is to have patches with the same amount of points. I am looking for a way to populate sparse point clouds to a given threshold without enlarging their outlines in python. I was planning to implement bilinear interpolation but I could not figure out how to adapt it in my case.

Is there any library or solution to handle this problem?

Looking forward to hearing your suggestions and helps! Many thanks in advance.

  • Please edit the question to limit it to a specific problem with enough detail to identify an adequate answer. – Community Jul 09 '22 at 00:48

1 Answers1

0

One way to do this is to randomly sample points from the patch until you have the desired number of points. This will preserve the overall distribution of points in the patch.

Soufian
  • 90
  • 4