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

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 1–25 of 1459 papers

TitleStatusHype
Unsupervised Part Discovery via Descriptor-Based Masked Image Restoration with Optimized ConstraintsCode0
LD-RPS: Zero-Shot Unified Image Restoration via Latent Diffusion Recurrent Posterior SamplingCode1
Double-Diffusion: Diffusion Conditioned Diffusion Probabilistic Model For Air Quality Prediction—0
EAMamba: Efficient All-Around Vision State Space Model for Image RestorationCode2
Elucidating and Endowing the Diffusion Training Paradigm for General Image Restoration—0
Wild refitting for black box prediction—0
TDiR: Transformer based Diffusion for Image Restoration Tasks—0
NAADA: A Noise-Aware Attention Denoising Autoencoder for Dental Panoramic Radiographs—0
Enhancing Image Restoration Transformer via Adaptive Translation Equivariance—0
Reversing Flow for Image Restoration—0
Visual-Instructed Degradation Diffusion for All-in-One Image RestorationCode1
MoiréXNet: Adaptive Multi-Scale Demoiréing with Linear Attention Test-Time Training and Truncated Flow Matching Prior—0
Optimization-Based Image Restoration under Implementation Constraints in Optical Analog Circuits—0
Exploring Diffusion with Test-Time Training on Efficient Image Restoration—0
Unsupervised Imaging Inverse Problems with Diffusion Distribution MatchingCode1
Exploiting the Exact Denoising Posterior Score in Training-Free Guidance of Diffusion Models—0
Text-Aware Image Restoration with Diffusion Models—0
M2Restore: Mixture-of-Experts-based Mamba-CNN Fusion Framework for All-in-One Image Restoration—0
Multi-Step Guided Diffusion for Image Restoration on Edge Devices: Toward Lightweight Perception in Embodied AI—0
UniRes: Universal Image Restoration for Complex Degradations—0
SeedVR2: One-Step Video Restoration via Diffusion Adversarial Post-Training—0
NTIRE 2025 Challenge on RAW Image Restoration and Super-Resolution—0
NTIRE 2025 the 2nd Restore Any Image Model (RAIM) in the Wild Challenge—0
Image Restoration Learning via Noisy Supervision in the Fourier Domain—0
Boosting All-in-One Image Restoration via Self-Improved Privilege LearningCode1
Show:102550
← PrevPage 1 of 59Next →
All datasetsCDD-11UHDM.

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1OneRestoreAverage PSNR (dB)28.72—Unverified
2SRUDCAverage PSNR (dB)27.64—Unverified
3RestormerAverage PSNR (dB)26.99—Unverified
4WGWSNetAverage PSNR (dB)26.96—Unverified
5DGUNetAverage PSNR (dB)26.92—Unverified
6OKNetAverage PSNR (dB)26.33—Unverified
7MIRNetAverage PSNR (dB)25.97—Unverified
8PromptIRAverage PSNR (dB)25.9—Unverified
9MPRNetAverage PSNR (dB)25.47—Unverified
10MIRNetv2Average PSNR (dB)25.37—Unverified
#ModelMetricClaimedVerifiedStatus
1ESDNet-LPSNR22.42—Unverified
2ESDNetPSNR22.12—Unverified
#ModelMetricClaimedVerifiedStatus
1730L37—Unverified