TROVE Feature Detection for Online Pose Recovery by Binocular Cameras
This provides a promising localization method for unmanned robots in man-made environments, though it appears incremental as it builds on existing feature detection approaches.
The paper tackles real-time 6-DoF ego-state estimation by introducing TROVE features from man-made structures, achieving 60 Hz with accuracies of 0.3 degrees for attitude and 2 cm for position in indoor environments.
This paper proposes a new and efficient method to estimate 6-DoF ego-states: attitudes and positions in real time. The proposed method extract information of ego-states by observing a feature called "TROVE" (Three Rays and One VErtex). TROVE features are projected from structures that are ubiquitous on man-made constructions and objects. The proposed method does not search for conventional corner-type features nor use Perspective-n-Point (PnP) methods, and it achieves a real-time estimation of attitudes and positions up to 60 Hz. The accuracy of attitude estimates can reach 0.3 degrees and that of position estimates can reach 2 cm in an indoor environment. The result shows a promising approach for unmanned robots to localize in an environment that is rich in man-made structures.