GenZ-ICP enhances LiDAR odometry by introducing an adaptive weighting scheme for ICP registration, improving robustness in challenging environments like tunnels and open fields. It builds on KISS-ICP with Python and ROS integration.
OverlapNet uses Siamese networks on 2D range images from 3D LiDAR to detect loop closures by predicting overlap and relative yaw angle simultaneously. Practical demos included.
Boxer3D is a native Swift iOS app that runs real-time 3D object detection on iPhone using LiDAR and a YOLO11n+BoxerNet pipeline accelerated by ONNX Runtime and CoreML.
Super-LIO improves LiDAR-Inertial Odometry with a compact mapping strategy that speeds up correspondence search by 1.2-4x. It supports ROS1/2 and targets Livox Mid-360 sensors.
