Abstract
This paper presents an efficient and flexible solution for camera autocalibration from N≥3 views, given image correspondences and zero (or known) skew only. The knowledge is not required on camera motion, 3D information, scene, or internal constraints. Our method is essentially based only on the fundamental matrices and its main virtues are threefold. Firstly, it is shown that, in the center-oriented metric coordinates, the focal length and aspect ratio can be estimated independent of considerable principle point shift (PPs). Thus, our method includes recursive steps: estimating focal length and aspect ratio and then calculating the PPs. Secondly, the optimal geometric constraints are selected for calibration by using error propagation analyses. Thirdly, the Levenberg–Marquardt algorithm is adopted for the fast final refinement of four internal parameters. Our method is fast and efficient to derive a unique calibration. Besides, this method can be applied to calibrate the focal length from two views, without requiring the prior knowledge of PPs. Good performance of our method is evaluated and confirmed in both the simulation experiments and the practical tests.
Highlights
Camera calibration is an essential topic in photogrammetry and computer vision
This paper presents an efficient and flexible solution for camera autocalibration from N≥3 views, given image correspondences and zero skew only
We presented an efficient method for camera autocalibration from N≥3 views, by given zero skew and image correspondences only
Summary
Camera calibration is an essential topic in photogrammetry and computer vision. Numerous works have been investigated in this subject in the last decades. We present an efficient and flexible solution for camera autocalibration from N≥3 views ( from N≥2 fundamental matrices), given image correspondences and zero (or known) skew only. Its main virtues are threefold, distinctive from previous methods It shows that, in the center-oriented metric coordinates, the focal length and aspect ratio can be precisely estimated independent of considerable PPs. in contrast to the conventional methods which reconstruct simultaneously all the unknown internal parameters, our method contains two recursive steps. This iteration is very fast since it is performed on only four unknown internal parameters Another significant characteristic of the present method is using center-oriented metric coordinate system. Since there are four unknown internal parameters in (1) and two independent constraints in (3), two fundamental matrices shall be possible to perform autocalibration
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
