Uncertainty-Aware Source-Free Adaptive Image Restoration with State Space Augmentation.
Saved in:
| Title: | Uncertainty-Aware Source-Free Adaptive Image Restoration with State Space Augmentation. |
|---|---|
| Authors: | Ai, Yuang1,2 (AUTHOR) shallowdream555@gmail.com, Cao, Jie1,2 (AUTHOR) jie.cao@cripac.ia.ac.cn, He, Ran1,2 (AUTHOR) rhe@nlpr.ia.ac.cn, Huang, Huaibo1,2 (AUTHOR) huaibo.huang@cripac.ia.ac.cn |
| Source: | International Journal of Computer Vision. Jun2026, Vol. 134 Issue 6, p1-22. 22p. |
| Subjects: | Data augmentation, Linear dynamical systems, Image reconstruction, Machine learning |
| Abstract: | Real-world image restoration faces considerable domain gaps, both between synthetic and real data and among various real datasets (e.g., images acquired from different camera devices). Unsupervised domain adaptation (UDA) is a powerful approach for bridging domain gaps by utilizing both source and target data. Given the privacy policies and transmission restrictions often associated with source data in practical scenarios, we propose a SOurce-free Domain Adaptation framework for Image Restoration (SODA-IR), allowing a source-trained model to adapt to a target domain using only unlabeled target data without target domain ground truth images. It employs the source-trained model to produce refined pseudo-labels, which are then used to facilitate teacher-student learning. To better utilize pseudo-labels, we introduce WaveMamba, a novel wavelet-based augmentation method built upon the advanced state space model, Mamba. WaveMamba can be seamlessly integrated with existing networks to generate efficacious augmented data implicitly. Leveraging Mamba's advantage of linear complexity in long sequence modeling, it enables efficient interactions between tokens at different spatial positions across various samples. Additionally, we propose an uncertainty-aware self-training mechanism to enhance the accuracy of pseudo-labels, where inaccurate predictions are rectified through uncertainty estimation. To assess our SODA-IR, we conduct comprehensive experiments across various real-world image restoration tasks, including image super-resolution, motion deblurring, deraining, and denoising. Without accessing source data, SODA-IR surpasses state-of-the-art UDA methods, while also being adaptable across both regression-based and diffusion-based models. [ABSTRACT FROM AUTHOR] |
| Copyright of International Journal of Computer Vision is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) | |
| Database: | Engineering Source |
| FullText | Text: Availability: 0 |
|---|---|
| Header | DbId: egs DbLabel: Engineering Source An: 193394316 AccessLevel: 6 PubType: Academic Journal PubTypeId: academicJournal PreciseRelevancyScore: 0 |
| IllustrationInfo | |
| Items | – Name: Title Label: Title Group: Ti Data: Uncertainty-Aware Source-Free Adaptive Image Restoration with State Space Augmentation. – Name: Author Label: Authors Group: Au Data: <searchLink fieldCode="AR" term="%22Ai%2C+Yuang%22">Ai, Yuang</searchLink><relatesTo>1,2</relatesTo> (AUTHOR)<i> shallowdream555@gmail.com</i><br /><searchLink fieldCode="AR" term="%22Cao%2C+Jie%22">Cao, Jie</searchLink><relatesTo>1,2</relatesTo> (AUTHOR)<i> jie.cao@cripac.ia.ac.cn</i><br /><searchLink fieldCode="AR" term="%22He%2C+Ran%22">He, Ran</searchLink><relatesTo>1,2</relatesTo> (AUTHOR)<i> rhe@nlpr.ia.ac.cn</i><br /><searchLink fieldCode="AR" term="%22Huang%2C+Huaibo%22">Huang, Huaibo</searchLink><relatesTo>1,2</relatesTo> (AUTHOR)<i> huaibo.huang@cripac.ia.ac.cn</i> – Name: TitleSource Label: Source Group: Src Data: <searchLink fieldCode="JN" term="%22International+Journal+of+Computer+Vision%22">International Journal of Computer Vision</searchLink>. Jun2026, Vol. 134 Issue 6, p1-22. 22p. – Name: Subject Label: Subjects Group: Su Data: <searchLink fieldCode="DE" term="%22Data+augmentation%22">Data augmentation</searchLink><br /><searchLink fieldCode="DE" term="%22Linear+dynamical+systems%22">Linear dynamical systems</searchLink><br /><searchLink fieldCode="DE" term="%22Image+reconstruction%22">Image reconstruction</searchLink><br /><searchLink fieldCode="DE" term="%22Machine+learning%22">Machine learning</searchLink> – Name: Abstract Label: Abstract Group: Ab Data: Real-world image restoration faces considerable domain gaps, both between synthetic and real data and among various real datasets (e.g., images acquired from different camera devices). Unsupervised domain adaptation (UDA) is a powerful approach for bridging domain gaps by utilizing both source and target data. Given the privacy policies and transmission restrictions often associated with source data in practical scenarios, we propose a SOurce-free Domain Adaptation framework for Image Restoration (SODA-IR), allowing a source-trained model to adapt to a target domain using only unlabeled target data without target domain ground truth images. It employs the source-trained model to produce refined pseudo-labels, which are then used to facilitate teacher-student learning. To better utilize pseudo-labels, we introduce WaveMamba, a novel wavelet-based augmentation method built upon the advanced state space model, Mamba. WaveMamba can be seamlessly integrated with existing networks to generate efficacious augmented data implicitly. Leveraging Mamba's advantage of linear complexity in long sequence modeling, it enables efficient interactions between tokens at different spatial positions across various samples. Additionally, we propose an uncertainty-aware self-training mechanism to enhance the accuracy of pseudo-labels, where inaccurate predictions are rectified through uncertainty estimation. To assess our SODA-IR, we conduct comprehensive experiments across various real-world image restoration tasks, including image super-resolution, motion deblurring, deraining, and denoising. Without accessing source data, SODA-IR surpasses state-of-the-art UDA methods, while also being adaptable across both regression-based and diffusion-based models. [ABSTRACT FROM AUTHOR] – Name: AbstractSuppliedCopyright Label: Group: Ab Data: <i>Copyright of International Journal of Computer Vision is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract.</i> (Copyright applies to all Abstracts.) |
| PLink | https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=egs&AN=193394316 |
| RecordInfo | BibRecord: BibEntity: Identifiers: – Type: doi Value: 10.1007/s11263-026-02848-6 Languages: – Code: eng Text: English PhysicalDescription: Pagination: PageCount: 22 StartPage: 1 Subjects: – SubjectFull: Data augmentation Type: general – SubjectFull: Linear dynamical systems Type: general – SubjectFull: Image reconstruction Type: general – SubjectFull: Machine learning Type: general Titles: – TitleFull: Uncertainty-Aware Source-Free Adaptive Image Restoration with State Space Augmentation. Type: main BibRelationships: HasContributorRelationships: – PersonEntity: Name: NameFull: Ai, Yuang – PersonEntity: Name: NameFull: Cao, Jie – PersonEntity: Name: NameFull: He, Ran – PersonEntity: Name: NameFull: Huang, Huaibo IsPartOfRelationships: – BibEntity: Dates: – D: 01 M: 06 Text: Jun2026 Type: published Y: 2026 Identifiers: – Type: issn-print Value: 09205691 Numbering: – Type: volume Value: 134 – Type: issue Value: 6 Titles: – TitleFull: International Journal of Computer Vision Type: main |
| ResultId | 1 |
