Automatic Calibration Method for High Resolution LiDAR and Fisheye Camera

With the rapid development of autonomous vehicles, researchers and engineers have brought LiDAR sensors to the academic and the industrial attention since the LiDAR sensor can not only provide accurate range information but also generate various geometrical structures. Importantly, the combination of cameras and LiDAR sensors, as a typical example, benefits the perception a lot by offering rich and precise sensing data. In this paper, we propose a novel pipeline that can estimate an accurate rigid-body transformation when calibrating the extrinsic parameter between a fisheye camera and a LiDAR sensor. This system uses a checkerboard as the target to build the correspondence between these two sensors. By using the extracted 2D feature points and 3D feature points from a camera and a LiDAR sensor, the extrinsic parameter is calculated via a reformulated approach. In addition, we demonstrate the result of the proposed method and conclude this method lastly.