LAUSR

Monocular visual localization plays a pivotal role in advanced driver assistance systems and autonomous driving by estimating a vehicle’s ego motion from a single pinhole camera. Nevertheless, conventional monocular visual odometry (MVO) encounters challenges in scale estimation due to the absence of depth information during projection. Previous methodologies, whether rooted in physical constraints or deep learning paradigms, contend with issues related to computational complexity and the management of dynamic objects. This study extends our prior research, presenting innovative strategies for ego-motion estimation and ground points selection. Striving for a nuanced equilibrium between computational efficiency and precision, we propose a hybrid method that leverages the SegNeXt model for real-time applications, encompassing both ego-motion estimation and ground point selection. Our methodology incorporates dynamic-object masks to eliminate unstable features and employs ground-plane masks for meticulous triangulation. Furthermore, we exploit geometry constraint to delineate road regions for scale recovery (SR). The integration of this approach with the monocular version of ORB-SLAM3 culminates in the accurate estimation of a road model (RM), a pivotal component in our SR process. Rigorous experiments, conducted on the KITTI dataset, systematically compare our method with existing MVO algorithms and contemporary SR methodologies. The results undeniably confirm the superior effectiveness of our approach, surpassing state-of-the-art visual odometry (VO) algorithms. Our source code is available at: https://github.com/bFr0zNq/MVOSegScale