Constant-time monocular self-calibration

This paper describes an extensible framework for real-time self-calibration of cameras in the simultaneous mapping and localization (SLAM) setting. The system is demonstrated to calibrate both pinhole and fish-eye camera models from unknown initial parameters while seamlessly solving the maximum likelihood online SLAM problem in real-time. Self-calibration is performed by tracking image features, and requires no predetermined calibration target. By automatically identifying and using only those portions of the sequence that contain useful information for the purpose of calibration the system achieves accurate results incrementally and in constant-time vs. the number of images. Furthermore, no special initialization movements are necessary. Parameters estimated by the framework are shown to closely match the batch solution as well as offline calibration values, but are computed live in constant-time. By not rolling information into an assumed prior distribution, the system avoids inconsistencies caused by early linearization - a problem that limits filtering techniques. The system is evaluated with experimental data and shown to be accurate vs. both the offline and batch calibration estimates.