true-real-michael/python-plane-ransac

Parallel RANSAC for plane detection for multiple point clouds using Python and CUDA

Python Plane RANSAC with CUDA

This is an implementation of the RANSAC algorithm for plane detection in point clouds. The algorithm is implemented in Python using Numba.

This implementation can can be used to segment planes from multiple point clouds in parallel using CUDA. The hypotheses are tested in parallel as well.

Installation

To run the exmples, clone the repository

git clone https://github.com/true-real-michael/python-plane-ransac.git

To use in your own code, install via pip

pip install git+https://github.com/true-real-michael/python-plane-ransac.git

Usage

See the examples for plane detection and outlier removal in the examples folder.

If GitHub fails to render the notebooks, you can view the examples via Jupyter NBViewer: plane detection and outlier removal.