Det neighbours within radius of each point , radius changes per point relating to c value. Sk-learn to slow anything faster?

Question

I am working on a point cloud simplification algorithm that preserves features.

In the one aspect I am searching through each xyzc point, at each point i search the nearest neighbours within a radius - this radius is related to the points c value.

I have been using sk-learn nearest neighbours but it is super slow!!!

Is there any other library I can use that is much faster? Thanks

Answer 1

If exact nearest neighbors search is too slow you should probably try approximate nearest neighbors search instead, for instance using the flann library.