In this paper, we present an Adaptive Feature Optimization Strategy as a novel frontend feature selection strategy which can be applied in the most direct method-based Simultaneous Localization And Mapping (SLAM) system for high accuracy, efficiency and robustness compared to the original SLAM system. It chooses adaptively the ORB points or Direct Sparse Odometry (DSO)-based points for tracking depending on in which scenario the cameras situate. Our evaluation on public datasets presents that the SLAM system integrated with our strategy outperforms the state-of-the-art which significantly reduces the processing time for each frame while retains the tracking accuracy.
An Adaptive Feature Optimization Strategy for Direct Visual Odometry
Lect. Notes Electrical Eng.
International Conference on Autonomous Unmanned Systems ; 2021 ; Changsha, China September 24, 2021 - September 26, 2021
Proceedings of 2021 International Conference on Autonomous Unmanned Systems (ICAUS 2021) ; Chapter : 286 ; 2919-2930
2022-03-18
12 pages
Article/Chapter (Book)
Electronic Resource
English
An Adaptive Feature Optimization Strategy for Direct Visual Odometry
British Library Conference Proceedings | 2022
|IEEE | 2022
|