The geometrical calibration of a high-definition camera rig is an important step for 3D film making and computer vision applications. Due to the large amount of image data in high-definition, maintaining execution speeds appropriate for on-set, on-line adjustment procedures is one of the biggest challenges for machine vision based calibration methods. Our aims are to provide a low-cost, fast and accurate system to calibrate both the intrinsic and extrinsic parameters of a stereo camera rig. We first propose a novel calibration target that we call marker chessboard to speed up the corner detection. Then we develop an automatic key frame selection algorithm to optimize frames used in calibration. We also propose a bundle adjustment method to overcome the geometrical inaccuracy of the chessboard. Finally we introduce an online stereo camera calibration system based on the above improvements.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.