SOTAVerified

Unified Image Restoration

Using a single model to restore inputs with different degradation types.

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 1–10 of 17 papers

TitleStatusHype
Photo-Realistic Image Restoration in the Wild with Controlled Vision-Language ModelsCode4
Degradation-Aware Residual-Conditioned Optimal Transport for Unified Image RestorationCode3
CRNet: A Detail-Preserving Network for Unified Image Restoration and Enhancement TaskCode2
Controlling Vision-Language Models for Multi-Task Image RestorationCode2
Boosting All-in-One Image Restoration via Self-Improved Privilege LearningCode1
Neural Degradation Representation Learning for All-In-One Image RestorationCode1
LD-RPS: Zero-Shot Unified Image Restoration via Latent Diffusion Recurrent Posterior SamplingCode1
ProRes: Exploring Degradation-aware Visual Prompt for Universal Image RestorationCode1
Unified Image Restoration and Enhancement: Degradation Calibrated Cycle Reconstruction Diffusion ModelCode1
Beyond Degradation Redundancy: Contrastive Prompt Learning for All-in-One Image RestorationCode1
Show:102550
← PrevPage 1 of 2Next →
All datasetsBSD68 sigma25GoProLOLRain100LRESIDE

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1DA-RCOTAverage PSNR (dB)31.23—Unverified
#ModelMetricClaimedVerifiedStatus
1DA-RCOTAverage PSNR (dB)28.68—Unverified
#ModelMetricClaimedVerifiedStatus
1DA-RCOTAverage PSNR (dB)23.25—Unverified
#ModelMetricClaimedVerifiedStatus
1DA-RCOTAverage PSNR (dB)38.36—Unverified
#ModelMetricClaimedVerifiedStatus
1DA-RCOTAverage PSNR (dB)31.26—Unverified