Extrinsic calibration of camera and 3D laser sensor system

Abstract

Robots are typically equipped with multiple complementary sensors such as cameras and laser range finders. Camera generally provides dense 2D information while range sensors give sparse and accurate depth information in the form of a set of 3D points. In order to represent the different data sources in a common coordinate system, extrinsic calibration is needed. This paper presents a pipeline for extrinsic calibration a zed setero camera with Velodyne LiDAR puck using a novel self-made 3D marker whose edges can be robustly detected in the image and 3d point cloud. Our approach first estimate the large sensor displacement using just a single frame. then we optimize the coarse results by finding the best align of edges in order to obtain a more accurate calibration. Finally, the ratio of the 3D points correctly projected onto proper image segments is used to evaluate the accuracy of calibration.