Camera calibration method for solid spheres based on triangular primitives

Abstract

A new camera calibration method based on triangular primitives for solid grid sphere objects are proposed in this work to address the contradiction among the number of cameras, calibration accuracy, and algorithm stability in the current camera calibration process. In the proposed method, a sphere is used as the base, and the surface coverage curve of the sphere forms a calibration object as the grid based on triangular primitives. We use the absolute quadratic curve and vanishing point to calculate the intrinsic and extrinsic parameters of the camera, respectively. The proposed method can simplify the element correspondence in the calibration and calibrate multiple cameras in different positions simultaneously. Experimental results show that the new 3D calibrator exceeds the calibration accuracy and stability of traditional algorithms. Moreover, the comprehensive evaluation of the algorithm is improved significantly.