3D feature points calibration method for depth-camera

The depth-camera calibration is the basis of application of the depth camera in 3D vision fields, while the depth-camera calibration still presents numerous challenges, for example accuracy and ease of use. In this paper, we propose a novel depth-camera calibration method: the depth-camera calibration method based on 3D feature points (DCCM-3D), DCCM-3D calibrates the depth-camera by directly using depth-images instead of using 2D images. Firstly, we build the depth information transform model of the world coordinate system to the pixel coordinate system by adding the depth information value to the 2D pixel coordinate system. Secondly, two calibration principles for the depth-camera are putting forward: point calibration principle and image calibration principle, which shows that the depth-camera calibration can be finished by obtaining two depth-images and the coordinates of four 3D feature points in each depth-image.