A VARIANT OF LSD-SLAM CAPABLE OF PROCESSING HIGH-SPEED LOW-FRAMERATE MONOCULAR DATASETS
Keywords: monocular visual odometry, filter-based visual odometry, Augmented Reality
Abstract. We develop a new variant of LSD-SLAM, called C-LSD-SLAM, which is capable of performing monocular tracking and mapping in high-speed low-framerate situations such as those of the KITTI datasets. The methods used here are robust against the influence of erronously triangulated points near the epipolar direction, which otherwise causes tracking divergence.