Transformer Based Visual Inertial Odometry

Sicheng Fei, Jingfeng Li, Lei Li, Jie Liang, Jinwen Hu, Dingwen Zhang, Junwei Han

科研成果: 书/报告/会议事项章节会议稿件同行评审

摘要

Visual inertial odometry (VIO) is a sensor fusion technology used for positioning and navigation. It combines visual sensor and inertial sensor information to estimate the movement and location of the UAV in real time. In recent years deep learning based approaches VIO have shown outstanding performance than traditional geometric methods. However, VIO tasks usually need to capture long-distance feature dependencies to ensure the continuity and consistency of camera motion trajectories in time series. In this study, we introduce a new end to end transformer based VIO framework, named VIO-former, to enable the model to better understand motion features in video sequences. Comprehensive quantitative and qualitative evaluation is conducted on KITTI datasets to test our method. The experimental results shows that our approach can achieve superior performance when compared with the existing methods.

源语言英语
主期刊名Advances in Guidance, Navigation and Control - Proceedings of 2024 International Conference on Guidance, Navigation and Control Volume 17
编辑Liang Yan, Haibin Duan, Yimin Deng
出版商Springer Science and Business Media Deutschland GmbH
567-575
页数9
ISBN(印刷版)9789819622634
DOI
出版状态已出版 - 2025
活动International Conference on Guidance, Navigation and Control, ICGNC 2024 - Changsha, 中国
期限: 9 8月 202411 8月 2024

出版系列

姓名Lecture Notes in Electrical Engineering
1353 LNEE
ISSN(印刷版)1876-1100
ISSN(电子版)1876-1119

会议

会议International Conference on Guidance, Navigation and Control, ICGNC 2024
国家/地区中国
Changsha
时期9/08/2411/08/24

指纹

探究 'Transformer Based Visual Inertial Odometry' 的科研主题。它们共同构成独一无二的指纹。

引用此