Gladkova M, Wang R, Zeller N, Cremers D (2021)
Publication Type: Conference contribution
Publication year: 2021
Publisher: Institute of Electrical and Electronics Engineers Inc.
Book Volume: 2021-May
Pages Range: 9608-9614
Conference Proceedings Title: Proceedings - IEEE International Conference on Robotics and Automation
Event location: Xi'an, CHN
ISBN: 9781728190778
DOI: 10.1109/ICRA48506.2021.9561217
In this paper we propose a framework for integrating map-based relocalization into online direct visual odometry. To achieve map-based relocalization for direct methods, we integrate image features into Direct Sparse Odometry (DSO) and rely on feature matching to associate online visual odometry (VO) with a previously built map. The integration of the relocalization poses is threefold. Firstly, they are incorporated as pose priors in the direct image alignment of the front-end tracking. Secondly, they are tightly integrated into the back-end bundle adjustment. Thirdly, an online fusion module is further proposed to combine relative VO poses and global relocalization poses in a pose graph to estimate keyframe-wise smooth and globally accurate poses. We evaluate our method on two multi-weather datasets showing the benefits of integrating different handcrafted and learned features and demonstrating promising improvements on camera tracking accuracy.
APA:
Gladkova, M., Wang, R., Zeller, N., & Cremers, D. (2021). Tight Integration of Feature-based Relocalization in Monocular Direct Visual Odometry. In Proceedings - IEEE International Conference on Robotics and Automation (pp. 9608-9614). Xi'an, CHN: Institute of Electrical and Electronics Engineers Inc..
MLA:
Gladkova, Mariia, et al. "Tight Integration of Feature-based Relocalization in Monocular Direct Visual Odometry." Proceedings of the 2021 IEEE International Conference on Robotics and Automation, ICRA 2021, Xi'an, CHN Institute of Electrical and Electronics Engineers Inc., 2021. 9608-9614.
BibTeX: Download