LIDAR
Sensor for measuring distances by illuminating the target with laser light.
Contents
Datasets
The dataset is time-stamped and contains raw data from all the sensors, calibration values, pose trajectory, ground truth pose, and 3D maps. The data is Robot Operating System (ROS) compatible.
Navigation and localisation dataset for self driving cars and autonomous robots in Brno, Czechia.
A dataset designed to support autonomous vehicle perception tasks including 3D tracking and motion forecasting collected in Pittsburgh, Pennsylvania and Miami, Florida, USA.
The Boreas dataset was collected by driving a repeated route over the course of 1 year resulting in stark seasonal variations. In total, Boreas contains over 350km of driving data including several sequences with adverse weather conditions such as rain and heavy snow. The Boreas data-taking platform features a unique high-quality sensor suite with a 128-channel Velodyne Alpha Prime lidar, a 360-degree Navtech radar, and accurate ground truth poses obtained from an Applanix POSLV GPS/IMU.
Libraries
Popular highly parallel programming library, with numerous industrial and research use-cases.
Open3D library contanins 3D data processing and visualization algorithms. It is open-source and supports both C++ and Python.
PyTorch3d is a library for deep learning with 3D data written and maintained by the Facebook AI Research Computer Vision Team.
Kaolin is a PyTorch Library for Accelerating 3D Deep Learning Research written by NVIDIA Technologies for game and application developers.
3D plotting and mesh analysis through a streamlined interface for the Visualization Toolkit.
Pyntcloud is a Python 3 library for working with 3D point clouds leveraging the power of the Python scientific stack.
Algorithms
simpleICP C++ /Julia / Matlab / Octave / Python implementation.
libpointmatcher, a modular library implementing the ICP algorithm.
Semantic segmentation
Ground segmentation
ROS comapatible ground plane segmentation; a library for fitting planes to LIDAR.
Line fitting-based fast ground segmentation for horizontal 3D LiDAR data
Region-wise plane fitting-based robust and fast ground segmentation for 3D LiDAR data
Improved version of Patchwork. Patchwork++ provides pybinding as well for deep learning users
Simultaneous localization and mapping SLAM and LIDAR-based odometry and or mapping LOAM
Remove, then Revert: Static Point cloud Map Construction using Multiresolution Range Images.
KISS-SLAM is a simple, robust, and accurate 3D LiDAR SLAM system that just works.
Object detection and object tracking
LIDAR-other-sensor calibration
Simulators
Unreal Engine based simulator for automotive applications. Compatible with Autoware, Baidu Apollo and ROS/ROS 2.
Unity Engine based simulator for automotive applications. Compatible with Autoware, Baidu Apollo and ROS/ROS 2. Note: LG has made the difficult decision to suspend active development of SVL Simulator.
Unity Engine based simulator for automotive applications, based on the suspended LGSVL simulator, but an active development. Compatible with Autoware, Baidu Apollo and ROS/ROS 2.
Related awesome
Others
Using $5 Google Cardboard to replicate Microsoft Hololens. Hosts the source code for research on scene color reconstruction.
A tool to turn point cloud data from aerial lidar scans into solid meshes for 3D printing.
Point cloud and image semantic segmentation editor by Hitachi Automotive And Industry Laboratory, point cloud annotator / labeling.