DiffLLE: Diffusion-based Domain Calibration for Weak Supervised Low-light Image Enhancement.
Saved in:
| Title: | DiffLLE: Diffusion-based Domain Calibration for Weak Supervised Low-light Image Enhancement. |
|---|---|
| Authors: | Yang, Shuzhou1 (AUTHOR) szyang@stu.pku.edu.cn, Zhang, Xuanyu1 (AUTHOR) xuanyuzhang21@stu.pku.edu.cn, Wang, Yinhuai1 (AUTHOR) yinhuai@stu.pku.edu.cn, Yu, Jiwen1 (AUTHOR) yujiwen@stu.pku.edu.cn, Wang, Yuhan1 (AUTHOR) yuhan.wang@stu.pku.edu.cn, Zhang, Jian1 (AUTHOR) zhangjian.sz@pku.edu.cn |
| Source: | International Journal of Computer Vision. May2025, Vol. 133 Issue 5, p2527-2546. 20p. |
| Subjects: | Image intensifiers, Artificial intelligence, Image processing, Calibration, Distillation |
| Abstract: | Existing weak supervised low-light image enhancement methods lack enough effectiveness and generalization in practical applications. We suppose this is because of the absence of explicit supervision and the inherent gap between real-world low-light domain and the training low-light domain. For example, low-light datasets are well-designed, but real-world night scenes are plagued with sophisticated interference such as noise, artifacts, and extreme lighting conditions. In this paper, we develop Diffusion-based domain calibration to realize more robust and effective weak supervised Low-Light Enhancement, called DiffLLE. Since the diffusion model performs impressive denoising capability and has been trained on massive clean images, we adopt it to bridge the gap between the real low-light domain and training degradation domain, while providing efficient priors of real-world content for weak supervised models. Specifically, we adopt a naive weak supervised enhancement algorithm to realize preliminary restoration and design two zero-shot plug-and-play modules based on diffusion model to improve generalization and effectiveness. The Diffusion-guided Degradation Calibration (DDC) module narrows the gap between real-world and training low-light degradation through diffusion-based domain calibration and a lightness enhancement curve, which makes the enhancement model perform robustly even in sophisticated wild degradation. Due to the limited enhancement effect of the weak supervised model, we further develop the Fine-grained Target domain Distillation (FTD) module to find a more visual-friendly solution space. It exploits the priors of the pre-trained diffusion model to generate pseudo-references, which shrinks the preliminary restored results from a coarse normal-light domain to a finer high-quality clean field, addressing the lack of strong explicit supervision for weak supervised methods. Benefiting from these, our approach even outperforms some supervised methods by using only a simple weak supervised baseline. Extensive experiments demonstrate the superior effectiveness of the proposed DiffLLE, especially in real-world dark scenes. [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 |
|
Full text is not displayed to guests.
Login for full access.
|
|
| FullText | Links: – Type: pdflink Text: Availability: 1 |
|---|---|
| Header | DbId: egs DbLabel: Engineering Source An: 184555880 AccessLevel: 6 PubType: Academic Journal PubTypeId: academicJournal PreciseRelevancyScore: 0 |
| IllustrationInfo | |
| Items | – Name: Title Label: Title Group: Ti Data: DiffLLE: Diffusion-based Domain Calibration for Weak Supervised Low-light Image Enhancement. – Name: Author Label: Authors Group: Au Data: <searchLink fieldCode="AR" term="%22Yang%2C+Shuzhou%22">Yang, Shuzhou</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> szyang@stu.pku.edu.cn</i><br /><searchLink fieldCode="AR" term="%22Zhang%2C+Xuanyu%22">Zhang, Xuanyu</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> xuanyuzhang21@stu.pku.edu.cn</i><br /><searchLink fieldCode="AR" term="%22Wang%2C+Yinhuai%22">Wang, Yinhuai</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> yinhuai@stu.pku.edu.cn</i><br /><searchLink fieldCode="AR" term="%22Yu%2C+Jiwen%22">Yu, Jiwen</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> yujiwen@stu.pku.edu.cn</i><br /><searchLink fieldCode="AR" term="%22Wang%2C+Yuhan%22">Wang, Yuhan</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> yuhan.wang@stu.pku.edu.cn</i><br /><searchLink fieldCode="AR" term="%22Zhang%2C+Jian%22">Zhang, Jian</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> zhangjian.sz@pku.edu.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>. May2025, Vol. 133 Issue 5, p2527-2546. 20p. – Name: Subject Label: Subjects Group: Su Data: <searchLink fieldCode="DE" term="%22Image+intensifiers%22">Image intensifiers</searchLink><br /><searchLink fieldCode="DE" term="%22Artificial+intelligence%22">Artificial intelligence</searchLink><br /><searchLink fieldCode="DE" term="%22Image+processing%22">Image processing</searchLink><br /><searchLink fieldCode="DE" term="%22Calibration%22">Calibration</searchLink><br /><searchLink fieldCode="DE" term="%22Distillation%22">Distillation</searchLink> – Name: Abstract Label: Abstract Group: Ab Data: Existing weak supervised low-light image enhancement methods lack enough effectiveness and generalization in practical applications. We suppose this is because of the absence of explicit supervision and the inherent gap between real-world low-light domain and the training low-light domain. For example, low-light datasets are well-designed, but real-world night scenes are plagued with sophisticated interference such as noise, artifacts, and extreme lighting conditions. In this paper, we develop Diffusion-based domain calibration to realize more robust and effective weak supervised Low-Light Enhancement, called DiffLLE. Since the diffusion model performs impressive denoising capability and has been trained on massive clean images, we adopt it to bridge the gap between the real low-light domain and training degradation domain, while providing efficient priors of real-world content for weak supervised models. Specifically, we adopt a naive weak supervised enhancement algorithm to realize preliminary restoration and design two zero-shot plug-and-play modules based on diffusion model to improve generalization and effectiveness. The Diffusion-guided Degradation Calibration (DDC) module narrows the gap between real-world and training low-light degradation through diffusion-based domain calibration and a lightness enhancement curve, which makes the enhancement model perform robustly even in sophisticated wild degradation. Due to the limited enhancement effect of the weak supervised model, we further develop the Fine-grained Target domain Distillation (FTD) module to find a more visual-friendly solution space. It exploits the priors of the pre-trained diffusion model to generate pseudo-references, which shrinks the preliminary restored results from a coarse normal-light domain to a finer high-quality clean field, addressing the lack of strong explicit supervision for weak supervised methods. Benefiting from these, our approach even outperforms some supervised methods by using only a simple weak supervised baseline. Extensive experiments demonstrate the superior effectiveness of the proposed DiffLLE, especially in real-world dark scenes. [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=184555880 |
| RecordInfo | BibRecord: BibEntity: Identifiers: – Type: doi Value: 10.1007/s11263-024-02292-4 Languages: – Code: eng Text: English PhysicalDescription: Pagination: PageCount: 20 StartPage: 2527 Subjects: – SubjectFull: Image intensifiers Type: general – SubjectFull: Artificial intelligence Type: general – SubjectFull: Image processing Type: general – SubjectFull: Calibration Type: general – SubjectFull: Distillation Type: general Titles: – TitleFull: DiffLLE: Diffusion-based Domain Calibration for Weak Supervised Low-light Image Enhancement. Type: main BibRelationships: HasContributorRelationships: – PersonEntity: Name: NameFull: Yang, Shuzhou – PersonEntity: Name: NameFull: Zhang, Xuanyu – PersonEntity: Name: NameFull: Wang, Yinhuai – PersonEntity: Name: NameFull: Yu, Jiwen – PersonEntity: Name: NameFull: Wang, Yuhan – PersonEntity: Name: NameFull: Zhang, Jian IsPartOfRelationships: – BibEntity: Dates: – D: 01 M: 05 Text: May2025 Type: published Y: 2025 Identifiers: – Type: issn-print Value: 09205691 Numbering: – Type: volume Value: 133 – Type: issue Value: 5 Titles: – TitleFull: International Journal of Computer Vision Type: main |
| ResultId | 1 |
