SOTAVerified

Image Restoration

Image Restoration is a family of inverse problems for obtaining a high quality image from a corrupted input image. Corruption may occur due to the image-capture process (e.g., noise, lens blur), post-processing (e.g., JPEG compression), or photography in non-ideal conditions (e.g., haze, motion blur).

Source: Blind Image Restoration without Prior Knowledge

Papers

Showing 276300 of 1459 papers

TitleStatusHype
DeblurGAN-v2: Deblurring (Orders-of-Magnitude) Faster and BetterCode1
Exploiting Raw Images for Real-Scene Super-ResolutionCode1
Equipping Diffusion Models with Differentiable Spatial Entropy for Low-Light Image EnhancementCode1
EnlightenGAN: Deep Light Enhancement without Paired SupervisionCode1
Exploiting Deep Generative Prior for Versatile Image Restoration and ManipulationCode1
Exploring and Evaluating Image Restoration Potential in Dynamic ScenesCode1
Enhancing Intrinsic Adversarial Robustness via Feature Pyramid DecoderCode1
Data augmentation for deep learning based accelerated MRI reconstruction with limited dataCode1
Enhancing RAW-to-sRGB with Decoupled Style Structure in Fourier DomainCode1
Early Exit or Not: Resource-Efficient Blind Quality Enhancement for Compressed ImagesCode1
A Variational Perspective on Solving Inverse Problems with Diffusion ModelsCode1
EDPN: Enhanced Deep Pyramid Network for Blurry Image RestorationCode1
Deep Blind Video Super-resolutionCode1
Enlighten Anything: When Segment Anything Model Meets Low-Light Image EnhancementCode1
Exploring Resolution and Degradation Clues as Self-supervised Signal for Low Quality Object DetectionCode1
Generalizable Medical Image Segmentation via Random Amplitude Mixup and Domain-Specific Image RestorationCode1
DaLPSR: Leverage Degradation-Aligned Language Prompt for Real-World Image Super-ResolutionCode1
Dual Pixel Exploration: Simultaneous Depth Estimation and Image RestorationCode1
End-to-End Learning for Joint Image Demosaicing, Denoising and Super-ResolutionCode1
DRT: A Lightweight Single Image Deraining Recursive TransformerCode1
D2HNet: Joint Denoising and Deblurring with Hierarchical Network for Robust Night Image RestorationCode1
BANet: Blur-aware Attention Networks for Dynamic Scene DeblurringCode1
DeblurDiNAT: A Compact Model with Exceptional Generalization and Visual Fidelity on Unseen DomainsCode1
Dynamic Attentive Graph Learning for Image RestorationCode1
CycleISP: Real Image Restoration via Improved Data SynthesisCode1
Show:102550
← PrevPage 12 of 59Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1OneRestoreAverage PSNR (dB)28.72Unverified
2SRUDCAverage PSNR (dB)27.64Unverified
3RestormerAverage PSNR (dB)26.99Unverified
4WGWSNetAverage PSNR (dB)26.96Unverified
5DGUNetAverage PSNR (dB)26.92Unverified
6OKNetAverage PSNR (dB)26.33Unverified
7MIRNetAverage PSNR (dB)25.97Unverified
8PromptIRAverage PSNR (dB)25.9Unverified
9MPRNetAverage PSNR (dB)25.47Unverified
10MIRNetv2Average PSNR (dB)25.37Unverified
#ModelMetricClaimedVerifiedStatus
1ESDNet-LPSNR22.42Unverified
2ESDNetPSNR22.12Unverified
#ModelMetricClaimedVerifiedStatus
1730L37Unverified