Improve generalization for neural visual-SLAM with Bayes online learning
Robotica · 2025-05 · 43 (5) · 1660-1674
DOI: 10.1017/s0263574725000360 ↗Kurz gefasst
Fachartikel in der Zeitschrift Robotica, online erschienen am 31.03.2025. Verfasst von Jun Liu und 1 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 55 Literaturangaben.
Abstract (Originalsprache)
Among various deep learning-based SLAM systems, many exhibit low accuracy and inadequate generalization on non-training datasets. The deficiency in generalization ability can result in significant errors within SLAM systems during real-world applications, particularly in environments that diverge markedly from those represented in the training set. This paper presents a methodology to enhance the generalization capabilities of deep learning SLAM systems. It leverages their superior performance in feature extraction and introduces Exponential Moving Average (EMA) and Bayes online learning to improve generalization and localization accuracy in varied scenarios. Experimental validation, utilizing Absolute Trajectory Error (ATE) metrics on the dataset, has been conducted. The results demonstrate that this method effectively reduces errors by $20\%$ on the EuRoC dataset and by $35\%$ on the TUM dataset, respectively.
Zitieren
APA
Liu, J., & Deng, H. (2025). Improve generalization for neural visual-SLAM with Bayes online learning. Robotica, 43(5), 1660-1674. https://doi.org/10.1017/s0263574725000360
Vancouver
Liu J, Deng H. Improve generalization for neural visual-SLAM with Bayes online learning. Robotica. 2025;43(5):1660-1674. doi:10.1017/s0263574725000360
BibTeX
@article{Liu2025,
author = {Liu, Jun and Deng, Haihang},
title = {Improve generalization for neural visual-SLAM with Bayes online learning},
journal = {Robotica},
year = {2025},
volume = {43},
number = {5},
pages = {1660--1674},
doi = {10.1017/s0263574725000360}
}
EUVIMED