MIVI: multi-stage feature matching for infrared and visible image
MIVI: multi-stage feature matching for infrared and visible image
About this item
Full title
Author / Creator
Di, Yide , Liao, Yun , Zhu, Kaijun , Zhou, Hao , Zhang, Yijia , Duan, Qing , Liu, Junhui and Lu, Mingyu
Publisher
Berlin/Heidelberg: Springer Berlin Heidelberg
Journal title
Language
English
Formats
Publication information
Publisher
Berlin/Heidelberg: Springer Berlin Heidelberg
Subjects
More information
Scope and Contents
Contents
The matching of infrared and visible images has a wide range of applications across various fields. However, the large difference between these two types of images poses a significant challenge to achieving accurate feature matching. In this paper, we introduce a novel feature matching method for infrared and visible images, named MIVI. Our proposed multi-stage matching architecture enables the model to capture both fine local feature details and remote dependencies, while our novel composite loss function optimizes the model at each stage and significantly improves the matching accuracy. Qualitative and quantitative experiments demonstrate that MIVI outperforms other excellent algorithms in terms of accuracy. The code will be released at:
https://github.com/LiaoYun0x0/MIVI
....
Alternative Titles
Full title
MIVI: multi-stage feature matching for infrared and visible image
Authors, Artists and Contributors
Author / Creator
Identifiers
Primary Identifiers
Record Identifier
TN_cdi_proquest_journals_2926959554
Permalink
https://devfeature-collection.sl.nsw.gov.au/record/TN_cdi_proquest_journals_2926959554
Other Identifiers
ISSN
0178-2789
E-ISSN
1432-2315
DOI
10.1007/s00371-023-02889-9
