3D Vision & Point Cloud Processing

Modern library for 3D data processing with Python and C++ APIs. Core features include 3D data structures, processing algorithms, scene reconstruction, surface alignment, 3D visualization, and GPU acceleration. MIT licensed.

Standalone, large-scale open project for 2D/3D image and point cloud processing. Comprehensive algorithms for filtering, feature estimation, surface reconstruction, registration, model fitting, and segmentation. BSD licensed.

FAIR's library of reusable components for deep learning with 3D data. Provides efficient 3D operators, differentiable rendering, and mesh processing tools integrated with PyTorch. BSD licensed.

Real-Time Appearance-Based Mapping library for RGB-D, Stereo and LiDAR SLAM. Graph-based SLAM approach with incremental appearance-based loop closure detection for large-scale and long-term operation. BSD licensed.

Open source robotics manipulation framework for ROS 2. Motion planning, manipulation, 3D perception, kinematics, control, and navigation for robotic arms. BSD-3-Clause licensed.

Feed-forward 3D foundation model for transforming and reconstructing streaming 3D scenes. Apache 2.0 licensed.