Research Article
A Novel Cross-Resolution Image Alignment for Multi-camera System
@INPROCEEDINGS{10.1007/978-3-031-23902-1_1, author={Kuo Chen and Tianqi Zheng and Chenxing He and Yeru Wang}, title={A Novel Cross-Resolution Image Alignment for Multi-camera System}, proceedings={Mobile Multimedia Communications. 15th EAI International Conference, MobiMedia 2022, Virtual Event, July 22-24, 2022, Proceedings}, proceedings_a={MOBIMEDIA}, year={2023}, month={2}, keywords={Local feature matching Image alignment Multi-camera imaging}, doi={10.1007/978-3-031-23902-1_1} }- Kuo Chen
Tianqi Zheng
Chenxing He
Yeru Wang
Year: 2023
A Novel Cross-Resolution Image Alignment for Multi-camera System
MOBIMEDIA
Springer
DOI: 10.1007/978-3-031-23902-1_1
Abstract
In complicated computer vision tasks, the multi-camera system is more effective than a single camera owing to the image fusion of multiple cameras, in which the image alignment is the essential first step. Especially, the cross-resolution image alignment caused by focal length difference has been extensively studied. A usual solution is using pyramid based local feature matching to create the mapping between input images with high and low-resolution. However, this kind of algorithm has high time and space complexity and is not applicable to front-end equipment such as the multi-camera system. Therefore, this paper proposes a fast and novel cross-resolution image alignment method based on the approximate focal length difference, including the coarse feature matching in low-resolution and the matching model estimation in high-resolution. At last, two types of test experiments are carried out using the industrial camera and SLR camera respectively. And the experimental results show that the proposed method performs well for cross-resolution image alignment, which can be widely used for multi-camera system.
