Camera Calibration by a Single Image of Balls: From Conics to the Absolute Conic
Hirohisa Teramoto, Gang Xu · 2002
In this paper we propose a new flexible technique to calibrate camera’s intrinsic parameters from a single image of balls. Balls are projected onto the image as ellipses, or conics. Each conic provides two constraints on the intrinsic parameters. To estimate all five intrinsic parameters, we need three balls. Ball’s size is arbitrary. Everyone can find a few balls around. It is more flexible than other predesigned calibration objects. Only one image is used, and there is no correspondence problem involved. We propose an algorithm to estimate the camera intrinsic parameters and the relative size and position of each ball optimally and simultaneously from the boundary points of each ball. Experimental results on both synthetic and real images show that the algorithm is effective and robust.