Progressive alignment and interwoven composition network for image stitching
Abstract As one of the fundamental tasks in computer graphics and image processing, image stitching aims to combine multiple images with overlapping regions to generate a high-quality naturalness panorama. Most deep learning based image stitching methods suffer from unsatisfactory performance, becau...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Springer
2024-12-01
|
| Series: | Complex & Intelligent Systems |
| Subjects: | |
| Online Access: | https://doi.org/10.1007/s40747-024-01702-x |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832571207768080384 |
|---|---|
| author | Xiaoting Fan Long Sun Zhong Zhang Tariq S. Durrani |
| author_facet | Xiaoting Fan Long Sun Zhong Zhang Tariq S. Durrani |
| author_sort | Xiaoting Fan |
| collection | DOAJ |
| description | Abstract As one of the fundamental tasks in computer graphics and image processing, image stitching aims to combine multiple images with overlapping regions to generate a high-quality naturalness panorama. Most deep learning based image stitching methods suffer from unsatisfactory performance, because they neglect the cooperation relationship and complementary information between reference image and target image. To address these issues, we propose a progressive alignment and interwoven composition network (PAIC-Net) to produce satisfactory panorama images, which learns the cooperation relationship by a progressive homography alignment module and captures the complementary information by an interwoven image composition module. Specifically, a progressive homography alignment module is presented to align the input images, which progressively warps the reference and target images by focusing more on the combination of self-features and cooperation features. Then, an interwoven image composition module is presented to seamlessly fuse aligned image pairs, where the complementary information of one-view is captured to guide another-view in an interweaved way. Finally, an alignment loss and a composition loss are introduced to reduce alignment distortions and enhance seam consistency of the final image stitching results. Experimental results on benchmark datasets demonstrate that PAIC-Net outperforms state-of-the-art image stitching methods both quantitatively and qualitatively. |
| format | Article |
| id | doaj-art-f066a9788ce84b9fa7b0bfd3b557047c |
| institution | Kabale University |
| issn | 2199-4536 2198-6053 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | Springer |
| record_format | Article |
| series | Complex & Intelligent Systems |
| spelling | doaj-art-f066a9788ce84b9fa7b0bfd3b557047c2025-02-02T12:49:19ZengSpringerComplex & Intelligent Systems2199-45362198-60532024-12-0111111410.1007/s40747-024-01702-xProgressive alignment and interwoven composition network for image stitchingXiaoting Fan0Long Sun1Zhong Zhang2Tariq S. Durrani3Tianjin Key Laboratory of Wireless Mobile Communications and Power Transmission, Tianjin Normal UniversitySchool of Information and Engineering, Tianjin University of CommerceTianjin Key Laboratory of Wireless Mobile Communications and Power Transmission, Tianjin Normal UniversityDepartment of Electronic and Electrical Engineering, University of StrathclydeAbstract As one of the fundamental tasks in computer graphics and image processing, image stitching aims to combine multiple images with overlapping regions to generate a high-quality naturalness panorama. Most deep learning based image stitching methods suffer from unsatisfactory performance, because they neglect the cooperation relationship and complementary information between reference image and target image. To address these issues, we propose a progressive alignment and interwoven composition network (PAIC-Net) to produce satisfactory panorama images, which learns the cooperation relationship by a progressive homography alignment module and captures the complementary information by an interwoven image composition module. Specifically, a progressive homography alignment module is presented to align the input images, which progressively warps the reference and target images by focusing more on the combination of self-features and cooperation features. Then, an interwoven image composition module is presented to seamlessly fuse aligned image pairs, where the complementary information of one-view is captured to guide another-view in an interweaved way. Finally, an alignment loss and a composition loss are introduced to reduce alignment distortions and enhance seam consistency of the final image stitching results. Experimental results on benchmark datasets demonstrate that PAIC-Net outperforms state-of-the-art image stitching methods both quantitatively and qualitatively.https://doi.org/10.1007/s40747-024-01702-xImage stitchingDeep learningProgressive homography AlignmentInterwoven image composition |
| spellingShingle | Xiaoting Fan Long Sun Zhong Zhang Tariq S. Durrani Progressive alignment and interwoven composition network for image stitching Complex & Intelligent Systems Image stitching Deep learning Progressive homography Alignment Interwoven image composition |
| title | Progressive alignment and interwoven composition network for image stitching |
| title_full | Progressive alignment and interwoven composition network for image stitching |
| title_fullStr | Progressive alignment and interwoven composition network for image stitching |
| title_full_unstemmed | Progressive alignment and interwoven composition network for image stitching |
| title_short | Progressive alignment and interwoven composition network for image stitching |
| title_sort | progressive alignment and interwoven composition network for image stitching |
| topic | Image stitching Deep learning Progressive homography Alignment Interwoven image composition |
| url | https://doi.org/10.1007/s40747-024-01702-x |
| work_keys_str_mv | AT xiaotingfan progressivealignmentandinterwovencompositionnetworkforimagestitching AT longsun progressivealignmentandinterwovencompositionnetworkforimagestitching AT zhongzhang progressivealignmentandinterwovencompositionnetworkforimagestitching AT tariqsdurrani progressivealignmentandinterwovencompositionnetworkforimagestitching |