Omnidirectional camera system improves visual location accuracy indoors and outdoors
OmniVLoc: Topo-metric Visual Localization for Omnidirectional Cameras
Computer Vision and Pattern RecognitionRobotics
Summary
Figuring out exactly where you are by looking at pictures can be tricky, especially when lighting or what you see around you changes. The authors made a system called OmniVLoc that uses special all-around cameras to get a full view, helping it know exactly where it is even when things look different. They combined several smart methods to make the system very precise and good at ignoring mistakes. Their tests show this system works better than others at finding accurate locations.
What this means in practice
- •For autonomous vehicle developers: Use omnidirectional cameras combined with multi-stage verification to improve vehicle localization under varying lighting and scene conditions.
- •For robotics navigation teams: Implement a robust topo-metric localization system for robots operating in environments with changing visual appearances.
Authors
Zhuo Yu, Shuyang Zhang, Huihua Zhu, Wenjia Ouyang, Zuo Zuo, Jianhao Jiao, Yilong Zhu, Xiaopin Zhong, Zongze Wu
Abstract
Omnidirectional cameras provide surround-view information, significantly overcoming the field-of-view constraints for traditional visual localization systems. However, existing localization systems fail to fully exploit the characteristics of omnidirectional cameras, struggling to deliver accurate and robust localization performance amidst scene content variations and illumination-induced appearance changes. In this work, we propose OmniVLoc, a surround-view localization system based on a topo-metric framework. By integrating state-of-the-art global place recognition and monocular depth estimation methods, we introduce a multi-constrained visual localization paradigm. By solving a scale-aware pose graph and performing precise spherical bundle adjustment, we accurately recover the metric-level pose for each query frame.Furthermore, a hierarchical verification mechanism is designed to filter out erroneous outputs across different stages, effectively bolstering the overall system robustness. Extensive experiments on both public and our custom datasets validate the superior accuracy and reliability of OmniVLoc. To benefit the community, our source code and dataset will be made publicly available upon publication.