The normal calculated by main_wnnc.py is the zero vector

[dongshengLiu]

Hello author, thank you for your wonderful work！
The normals I calculated using lidar_pcd.xyz are zero vectors, and adjusting parameters or cropping the point cloud don't help. How can I fix this?

[jsnln]

Hi, thank you for your interest!

I don't have a machine at hand to run this. But with a quick look I don't think this example can be solved by any winding number based methods, because there's no inside/outside for this point cloud. Please see our discussion on the limitation. However, I will try to run this when I have the resources.

[dongshengLiu]

When I downsample the above point cloud to 10 million points uniformly, I can calculate the normal. And I also tested multiple point clouds, all with the same problem.