Swin-transformer for weak feature matching
Abstract Feature matching in computer vision is crucial but challenging in weakly textured scenes due to the lack of pattern repetition. We introduce the SwinMatcher feature matching method, aimed at addressing the issues of low matching quantity and poor matching precision in weakly textured scenes...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-01-01
|
| Series: | Scientific Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1038/s41598-025-87309-9 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832585917638901760 |
|---|---|
| author | Yuan Guo Wenpeng Li Ping Zhai |
| author_facet | Yuan Guo Wenpeng Li Ping Zhai |
| author_sort | Yuan Guo |
| collection | DOAJ |
| description | Abstract Feature matching in computer vision is crucial but challenging in weakly textured scenes due to the lack of pattern repetition. We introduce the SwinMatcher feature matching method, aimed at addressing the issues of low matching quantity and poor matching precision in weakly textured scenes. Given the inherently significant local characteristics of image features, we employ a local self-attention mechanism to learn from weakly textured areas, maximally preserving the features of weak textures. To address the issue of incorrect matches in scenes with repetitive patterns, we use a cross-attention and positional encoding mechanism to learn the correct matches of repetitive patterns in two scenes, achieving higher matching precision. We also introduce a matching optimization algorithm that calculates the spatial expected coordinates of local two-dimensional heat maps of correspondences to obtain the final sub-pixel level matches. Experiments indicate that, under identical training conditions, the SwinMatcher outperforms other standard methods in pose estimation, homography estimation, and visual localization. It exhibits strong robustness and superior matching in weakly textured areas, offering a new research direction for feature matching in weakly textured images. |
| format | Article |
| id | doaj-art-67c301ae47fd4861b13c438b0d42f77f |
| institution | Kabale University |
| issn | 2045-2322 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-67c301ae47fd4861b13c438b0d42f77f2025-01-26T12:23:37ZengNature PortfolioScientific Reports2045-23222025-01-0115111410.1038/s41598-025-87309-9Swin-transformer for weak feature matchingYuan Guo0Wenpeng Li1Ping Zhai2Department of Computer Science and Technology, Heilongjiang UniversitySchool of Computer and Control Engineering, Qiqihar UniversityDepartment of Computer Science and Technology, Qilu University of TechnologyAbstract Feature matching in computer vision is crucial but challenging in weakly textured scenes due to the lack of pattern repetition. We introduce the SwinMatcher feature matching method, aimed at addressing the issues of low matching quantity and poor matching precision in weakly textured scenes. Given the inherently significant local characteristics of image features, we employ a local self-attention mechanism to learn from weakly textured areas, maximally preserving the features of weak textures. To address the issue of incorrect matches in scenes with repetitive patterns, we use a cross-attention and positional encoding mechanism to learn the correct matches of repetitive patterns in two scenes, achieving higher matching precision. We also introduce a matching optimization algorithm that calculates the spatial expected coordinates of local two-dimensional heat maps of correspondences to obtain the final sub-pixel level matches. Experiments indicate that, under identical training conditions, the SwinMatcher outperforms other standard methods in pose estimation, homography estimation, and visual localization. It exhibits strong robustness and superior matching in weakly textured areas, offering a new research direction for feature matching in weakly textured images.https://doi.org/10.1038/s41598-025-87309-9Feature matchingDeep learningWeak textureTransformer |
| spellingShingle | Yuan Guo Wenpeng Li Ping Zhai Swin-transformer for weak feature matching Scientific Reports Feature matching Deep learning Weak texture Transformer |
| title | Swin-transformer for weak feature matching |
| title_full | Swin-transformer for weak feature matching |
| title_fullStr | Swin-transformer for weak feature matching |
| title_full_unstemmed | Swin-transformer for weak feature matching |
| title_short | Swin-transformer for weak feature matching |
| title_sort | swin transformer for weak feature matching |
| topic | Feature matching Deep learning Weak texture Transformer |
| url | https://doi.org/10.1038/s41598-025-87309-9 |
| work_keys_str_mv | AT yuanguo swintransformerforweakfeaturematching AT wenpengli swintransformerforweakfeaturematching AT pingzhai swintransformerforweakfeaturematching |