Joint Target-Less Intrinsic and Extrinsic Camera-LiDAR Calibration using Deep Point Correspondences
Researchers have developed a novel pipeline for calibrating cameras and LiDAR sensors without requiring physical targets. This method jointly estimates both the intrinsic parameters of the camera, including distortion, and the extrinsic parameters that define the relative position and orientation between the camera and LiDAR. The approach leverages deep learning for pixel-point correspondences and integrates this with a nonlinear optimization process to refine both sets of calibration parameters simultaneously. AI
IMPACT Introduces a more robust and automated method for sensor calibration, crucial for autonomous systems.