leomariga/pyRANSAC-3D

Python library for fitting 3D primitive shapes to point clouds with RANSAC. Detect planes, spheres, cylinders, cuboids, circles, and lines.

What it solves

It provides a way to identify and fit primitive geometric shapes within noisy 3D point cloud data. This is essential for tasks like 3D reconstruction, object tracking, and 3D SLAM (Simultaneous Localization and Mapping).

How it works

The library implements the Random Sample Consensus (RANSAC) method. It iteratively selects random subsets of points to estimate the parameters of a geometric model and identifies the points (inliers) that fit that model within a specified threshold, effectively filtering out noise and outliers.

Who it’s for

Developers and researchers working with 3D spatial data, computer vision, and robotics who need to extract structural primitives from point clouds.

Highlights

  • Supports fitting for a wide variety of shapes: planes, cylinders, cuboids, spheres, lines, circles, and points.
  • Simple API that allows fitting shapes using NumPy arrays.
  • Open-source implementation with comprehensive documentation.

Related

  • Project
    MIT-SPARK/KISS-Matcher

    A fast and robust point cloud registration library that aligns 3D point clouds to determine relative rotation and translation.

  • Project
    neka-nat/probreg

    A library for point cloud registration using probabilistic models, offering a more robust alternative to ICP for both rigid and non-rigid transformations.

  • Project
    ducha-aiki/pydegensac

    A Python wrapper for LO-RANSAC and DEGENSAC used to estimate homography and fundamental matrices from sparse image correspondences.

  • Project
    norlab-ulaval/libpointmatcher

    A modular C++ library with Python bindings that implements the Iterative Closest Point (ICP) algorithm for aligning 3D point clouds in robotics and computer vision.

  • Project
    koide3/gtsam_points

    A collection of GTSAM factors and optimizers specifically designed for range-based SLAM and 3D point cloud registration.