NSTL国家科技图书文献中心

1. RECIPROCAL TRANSFORMATION BASED CONVEX VARIATIONAL IMAGE RESTORATION WITH MULTIPLICATIVE NOISE NSTL国家科技图书文献中心

Gan, Yu | Liu, Zhifang ... - 《Inverse problems and imaging》 - 2025,19(2) - 297～333 - 共37页

摘要：Image restoration with multiplicative noise is | widely studied in image* processing, among which how to* | related to the image* variable in the model proposed by* | reciprocal transformation based convex restoration* (RTCr* | good restoration* results in terms of SNRs*

关键词： Image restoration | multiplicative noise | total variation | convex optimization | primal-dual algorithm | proximal alternating linearized minimization method

2. Gated normalization unit for image restoration NSTL国家科技图书文献中心

Qingyu Wang | Haitao Wang ... - 《Pattern analysis and applications》 - 2025,28(1) - 16.1～16.19 - 共19页

摘要：Image restoration has been an integral part of | potential for designing image* restoration algorithms* | image restoration tasks, we achieve 32.93 dB PSNR on | image processing research with the goal of converting | the adaptation of normalization methods in image

关键词： Image restoration | Image deblurring | Gating mechanisms | CNN

3. Underwater Image Restoration With UNet Model NSTL国家科技图书文献中心

JunGui Wang | JunWei Hsieh - 《International Workshop on Advanced Imaging Technology (IWAIT) 2025》 - International Workshop on Advanced Imaging Technology - 2025, - 135101I.1～135101I.6 - 共6页

摘要： for underwater image* restoration. Underwater images* | underwater image* restoration has shown significant progress* | model for underwater image* restoration which test on* | underwater images, researchers have proposed restoration | can automatically learn and adapt to the restoration

关键词： UnderWater image restoration | UNet | LSUI dataset

4. MetaWeather: Few-Shot Weather-Degraded Image Restoration NSTL国家科技图书文献中心

Youngrae Kim | Younggeol Cho ... - 《Computer Vision - ECCV 2024,Part LXXIII》 - European Conference on Computer Vision - 2025, - 206～222 - 共17页

摘要：-art multi-weather image* restoration methods. Code is | weather-degraded image restoration as a few-shot* | restoration approaches are limited in their applicability to | a query image, and learns to adapt to unseen | Real-world weather conditions are intricate

关键词： Image restoration | Few-shot learning | Domain adaptation

5. Efficient Cascaded Multiscale Adaptive Network for Image Restoration NSTL国家科技图书文献中心

Yichen Zhou | Pan Zhou ... - 《Computer Vision - ECCV 2024,Part XIX》 - European Conference on Computer Vision - 2025, - 92～110 - 共19页

摘要：Image restoration, encompassing tasks such as | different scales, critical for diverse image restoration | restoration tasks, e.g., image deblurring, denoising and | kernels across local image* regions to efficiently handle* | efficiency, achieving comparable or superior restoration

关键词： Image deblurring | Image denoising | Image super-resolution | Image restoration

6. SeBIR: Semantic-guided burst image restoration NSTL国家科技图书文献中心

Liu H. | Shao M.... - 《Neural Networks》 - 2025,181 - Article 106834～Article 106834 - 共12页

摘要：© 2024Burst image* restoration methods offer* | burst image* restoration incorporating the semantic* | designed to boost local consistency and image* quality* | -resolution, burst denoising, and burst low-light image | the possibility of recovering faithful scene details

关键词： Burst image restoration | Joint alignment strategy | Segment anything | Semantic-guided fusion

7. When Fast Fourier Transform Meets Transformer for Image Restoration NSTL国家科技图书文献中心

Xingyu Jiang | Xiuhui Zhang ... - 《Computer Vision - ECCV 2024,Part XLV》 - European Conference on Computer Vision - 2025, - 381～402 - 共22页

摘要： phenomena, existing image* restoration methods focus on the* | for kinds of restoration* tasks. Instead of exploring* | restoration priors. In this work, we first review various | . Based on this, we propose an efficient image | restoration framework, dubbed SFHformer, which incorporates

关键词： Image restoration | Frequency feature | Deep learning

8. Osmosis: RGBD Diffusion Prior for Underwater Image Restoration NSTL国家科技图书文献中心

Opher Bar Nathan | Deborah Levy ... - 《Computer Vision - ECCV 2024,Part LXII》 - European Conference on Computer Vision - 2025, - 302～319 - 共18页

摘要：Underwater image* restoration is a challenging* | . Diffusion priors have emerged as strong image restoration | guidance method based on the underwater image* formation* | method outperforms state-of-the-art baselines for image | restoration on very challenging scenes. Our code, models and

关键词： Diffusion models | Physics-Based computer vision | Underwater image restoration

9. Multispectral Image Restoration by Generalized Opponent Transformation Total Variation NSTL国家科技图书文献中心

Zhantao Ma | Michael K. Ng - 《SIAM Journal on Imaging Sciences》 - 2025,18(1) - 246～279 - 共34页

摘要： multispectral image* restoration. To evaluate the effectiveness* | performance of various image* processing tasks. Numerous* | instead of the original multispectral image* domain. Here* | to existing multispectral image* total variation* | Multispectral images contain light information

关键词： Image restoration | Total variation | Opponent transformation | Multispectral image

10. Efficient Diffusion Model for Image Restoration by Residual Shifting NSTL国家科技图书文献中心

Zongsheng Yue | Jianyi Wang ... - 《IEEE Transactions on Pattern Analysis and Machine Intelligence》 - 2025,47(1) - 116～130 - 共15页 - 被引量：1

摘要：While diffusion-based image* restoration (IR* | face restoration, and image* deblurring, even only* | , namely image* super-resolution, image inpainting, blind* | ) methods have achieved remarkable success, they are still | limited by the low inference speed attributed to the

摘要：Image restoration with multiplicative noise is | widely studied in image* processing, among which how to* | related to the image* variable in the model proposed by* | reciprocal transformation based convex restoration* (RTCr* | good restoration* results in terms of SNRs*

关键词： Image restoration | multiplicative noise | total variation | convex optimization | primal-dual algorithm | proximal alternating linearized minimization method

摘要：Image restoration has been an integral part of | potential for designing image* restoration algorithms* | image restoration tasks, we achieve 32.93 dB PSNR on | image processing research with the goal of converting | the adaptation of normalization methods in image

关键词： Image restoration | Image deblurring | Gating mechanisms | CNN

摘要： for underwater image* restoration. Underwater images* | underwater image* restoration has shown significant progress* | model for underwater image* restoration which test on* | underwater images, researchers have proposed restoration | can automatically learn and adapt to the restoration

关键词： UnderWater image restoration | UNet | LSUI dataset

摘要：-art multi-weather image* restoration methods. Code is | weather-degraded image restoration as a few-shot* | restoration approaches are limited in their applicability to | a query image, and learns to adapt to unseen | Real-world weather conditions are intricate

关键词： Image restoration | Few-shot learning | Domain adaptation

摘要：Image restoration, encompassing tasks such as | different scales, critical for diverse image restoration | restoration tasks, e.g., image deblurring, denoising and | kernels across local image* regions to efficiently handle* | efficiency, achieving comparable or superior restoration

关键词： Image deblurring | Image denoising | Image super-resolution | Image restoration

摘要：© 2024Burst image* restoration methods offer* | burst image* restoration incorporating the semantic* | designed to boost local consistency and image* quality* | -resolution, burst denoising, and burst low-light image | the possibility of recovering faithful scene details

关键词： Burst image restoration | Joint alignment strategy | Segment anything | Semantic-guided fusion

摘要： phenomena, existing image* restoration methods focus on the* | for kinds of restoration* tasks. Instead of exploring* | restoration priors. In this work, we first review various | . Based on this, we propose an efficient image | restoration framework, dubbed SFHformer, which incorporates

关键词： Image restoration | Frequency feature | Deep learning

摘要：Underwater image* restoration is a challenging* | . Diffusion priors have emerged as strong image restoration | guidance method based on the underwater image* formation* | method outperforms state-of-the-art baselines for image | restoration on very challenging scenes. Our code, models and

关键词： Diffusion models | Physics-Based computer vision | Underwater image restoration

摘要： multispectral image* restoration. To evaluate the effectiveness* | performance of various image* processing tasks. Numerous* | instead of the original multispectral image* domain. Here* | to existing multispectral image* total variation* | Multispectral images contain light information

关键词： Image restoration | Total variation | Opponent transformation | Multispectral image

摘要：While diffusion-based image* restoration (IR* | face restoration, and image* deblurring, even only* | , namely image* super-resolution, image inpainting, blind* | ) methods have achieved remarkable success, they are still | limited by the low inference speed attributed to the

关键词： Diffusion models | Image restoration | Noise | Degradation | Superresolution | Schedules | Image synthesis

4008-161-200 800-990-8900

国家科技图书文献中心

摘要：Image restoration with multiplicative noise is | widely studied in image processing, among which how to | related to the image variable in the model proposed by | reciprocal transformation based convex restoration (RTCr | good restoration results in terms of SNRs

关键词： Image restoration | multiplicative noise | total variation | convex optimization | primal-dual algorithm | proximal alternating linearized minimization method

摘要：Image restoration has been an integral part of | potential for designing image restoration algorithms | image restoration tasks, we achieve 32.93 dB PSNR on | image processing research with the goal of converting | the adaptation of normalization methods in image

关键词： Image restoration | Image deblurring | Gating mechanisms | CNN

摘要： for underwater image restoration. Underwater images | underwater image restoration has shown significant progress | model for underwater image restoration which test on | underwater images, researchers have proposed restoration | can automatically learn and adapt to the restoration

关键词： UnderWater image restoration | UNet | LSUI dataset

摘要：-art multi-weather image restoration methods. Code is | weather-degraded image restoration as a few-shot | restoration approaches are limited in their applicability to | a query image, and learns to adapt to unseen | Real-world weather conditions are intricate

关键词： Image restoration | Few-shot learning | Domain adaptation

摘要：Image restoration, encompassing tasks such as | different scales, critical for diverse image restoration | restoration tasks, e.g., image deblurring, denoising and | kernels across local image regions to efficiently handle | efficiency, achieving comparable or superior restoration

关键词： Image deblurring | Image denoising | Image super-resolution | Image restoration

摘要：© 2024Burst image restoration methods offer | burst image restoration incorporating the semantic | designed to boost local consistency and image quality | -resolution, burst denoising, and burst low-light image | the possibility of recovering faithful scene details

关键词： Burst image restoration | Joint alignment strategy | Segment anything | Semantic-guided fusion

摘要： phenomena, existing image restoration methods focus on the | for kinds of restoration tasks. Instead of exploring | restoration priors. In this work, we first review various | . Based on this, we propose an efficient image | restoration framework, dubbed SFHformer, which incorporates

关键词： Image restoration | Frequency feature | Deep learning

摘要：Underwater image restoration is a challenging | . Diffusion priors have emerged as strong image restoration | guidance method based on the underwater image formation | method outperforms state-of-the-art baselines for image | restoration on very challenging scenes. Our code, models and

关键词： Diffusion models | Physics-Based computer vision | Underwater image restoration

摘要： multispectral image restoration. To evaluate the effectiveness | performance of various image processing tasks. Numerous | instead of the original multispectral image domain. Here | to existing multispectral image total variation | Multispectral images contain light information

关键词： Image restoration | Total variation | Opponent transformation | Multispectral image

摘要：While diffusion-based image restoration (IR | face restoration, and image deblurring, even only | , namely image super-resolution, image inpainting, blind | ) methods have achieved remarkable success, they are still | limited by the low inference speed attributed to the

关键词： Diffusion models | Image restoration | Noise | Degradation | Superresolution | Schedules | Image synthesis

4008-161-200 800-990-8900

国家科技图书文献中心

摘要：Image restoration with multiplicative noise is | widely studied in image* processing, among which how to* | related to the image* variable in the model proposed by* | reciprocal transformation based convex restoration* (RTCr* | good restoration* results in terms of SNRs*

摘要：Image restoration has been an integral part of | potential for designing image* restoration algorithms* | image restoration tasks, we achieve 32.93 dB PSNR on | image processing research with the goal of converting | the adaptation of normalization methods in image

摘要： for underwater image* restoration. Underwater images* | underwater image* restoration has shown significant progress* | model for underwater image* restoration which test on* | underwater images, researchers have proposed restoration | can automatically learn and adapt to the restoration

摘要：-art multi-weather image* restoration methods. Code is | weather-degraded image restoration as a few-shot* | restoration approaches are limited in their applicability to | a query image, and learns to adapt to unseen | Real-world weather conditions are intricate

摘要：Image restoration, encompassing tasks such as | different scales, critical for diverse image restoration | restoration tasks, e.g., image deblurring, denoising and | kernels across local image* regions to efficiently handle* | efficiency, achieving comparable or superior restoration

摘要：© 2024Burst image* restoration methods offer* | burst image* restoration incorporating the semantic* | designed to boost local consistency and image* quality* | -resolution, burst denoising, and burst low-light image | the possibility of recovering faithful scene details

摘要： phenomena, existing image* restoration methods focus on the* | for kinds of restoration* tasks. Instead of exploring* | restoration priors. In this work, we first review various | . Based on this, we propose an efficient image | restoration framework, dubbed SFHformer, which incorporates

摘要：Underwater image* restoration is a challenging* | . Diffusion priors have emerged as strong image restoration | guidance method based on the underwater image* formation* | method outperforms state-of-the-art baselines for image | restoration on very challenging scenes. Our code, models and

摘要： multispectral image* restoration. To evaluate the effectiveness* | performance of various image* processing tasks. Numerous* | instead of the original multispectral image* domain. Here* | to existing multispectral image* total variation* | Multispectral images contain light information

摘要：While diffusion-based image* restoration (IR* | face restoration, and image* deblurring, even only* | , namely image* super-resolution, image inpainting, blind* | ) methods have achieved remarkable success, they are still | limited by the low inference speed attributed to the