End-to-end Learned Visual Odometry with Events and Frames

arXiv:2309.09947v2 Announce Type: replace
Abstract: Visual Odometry (VO) is crucial for autonomous robotic navigation, especially in GPS-denied environments like planetary terrains. To improve robustness, recent model-based VO systems have begun combining standard and event-based cameras. Event cameras excel in low-light and high-speed motion, while standard cameras provide dense and easier-to-track features, even in low-textured areas. However, the field of image- and event-based VO still predominantly relies on model-based methods and is yet to fully integrate recent image-only advancements leveraging end-to-end …

abstract arxiv autonomous begun cameras cs.cv environments event events excel features gps light low navigation robotic robustness speed standard systems type visual