$ show hku-mars/FAST-Calib
FAST-Calib is a ROS-based application for LiDAR-camera extrinsic calibration that uses a target-based approach with ArUco markers and circular holes. The system processes synchronized camera images and LiDAR point clouds to detect calibration target features in both modalities, then computes the extrinsic transformation between sensor frames. It supports both solid-state (Livox) and mechanical (Ouster, Hesai) LiDAR sensors through adaptive preprocessing pipelines. The application includes single-scene calibration for immediate results and multi-scene joint calibration for improved accuracy. Key components include modular LiDAR and camera processing pipelines, geometric validation for target detection, and coordinate transformation utilities.
The LLM identified 3 files worth highlighting for reuse:
Contains shared data structures (Params, Common::Point), utility functions for point cloud processing, geometric validation (Square class for rectangle verification), and coordinate transformation helpers. Demonstrates excellent separation of concerns and reusable design patterns for sensor fusion applications.
Handles loading and preprocessing of LiDAR data from rosbags with automatic detection of sensor type (solid-state vs mechanical). Abstracts away differences between Livox CustomMsg and standard PointCloud2 formats, providing a unified point cloud interface for downstream processing. Shows skill in handling diverse sensor data formats in ROS applications.
Implements the complete LiDAR-based calibration target detection pipeline including plane segmentation, alignment to 2D, boundary point detection, Euclidean clustering, and 2D circle fitting. Features adaptive algorithms for different LiDAR types and robust geometric validation for calibration target verification. Demonstrates sophisticated point cloud processing techniques for feature extraction.