SOTAVerified

Image Denoising

Image Denoising is a computer vision task that involves removing noise from an image. Noise can be introduced into an image during acquisition or processing, and can reduce image quality and make it difficult to interpret. Image denoising techniques aim to restore an image to its original quality by reducing or removing the noise, while preserving the important features of the image.

( Image credit: Wide Inference Network for Image Denoising via Learning Pixel-distribution Prior )

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 1–50 of 1220 papers

TitleStatusHype
DenoDet: Attention as Deformable Multi-Subspace Feature Denoising for Target Detection in SAR ImagesCode4
FSID: Fully Synthetic Image Denoising via Procedural Scene GenerationCode4
Simple Baselines for Image RestorationCode4
Posterior-Mean Rectified Flow: Towards Minimum MSE Photo-Realistic Image RestorationCode4
InstructIR: High-Quality Image Restoration Following Human InstructionsCode4
DiffBIR: Towards Blind Image Restoration with Generative Diffusion PriorCode4
Energy-Based Transformers are Scalable Learners and ThinkersCode4
SwinIR: Image Restoration Using Swin TransformerCode3
MAXIM: Multi-Axis MLP for Image ProcessingCode3
HAT: Hybrid Attention Transformer for Image RestorationCode3
Optimal Weighted Convolution for Classification and DenosingCode2
Rethinking Transformer-Based Blind-Spot Network for Self-Supervised Image DenoisingCode2
Improving Image Restoration by Revisiting Global Information AggregationCode2
Deep PCB To COCO ConvertorCode2
SUNet: Swin Transformer UNet for Image DenoisingCode2
Transfer CLIP for Generalizable Image DenoisingCode2
Noise Modeling in One Hour: Minimizing Preparation Efforts for Self-supervised Low-Light RAW Image DenoisingCode2
PromptIR: Prompting for All-in-One Blind Image RestorationCode2
Masked Image Training for Generalizable Deep Image DenoisingCode2
Hybrid Convolutional and Attention Network for Hyperspectral Image DenoisingCode2
Optimal Density Functions for Weighted Convolution in Learning ModelsCode2
Practical Blind Image Denoising via Swin-Conv-UNet and Data SynthesisCode2
Riemannian Optimization on Relaxed Indicator Matrix ManifoldCode2
SSUMamba: Spatial-Spectral Selective State Space Model for Hyperspectral Image DenoisingCode2
Beyond Text: Frozen Large Language Models in Visual Signal ComprehensionCode2
KBNet: Kernel Basis Network for Image RestorationCode2
Z*: Zero-shot Style Transfer via Attention ReweightingCode2
Real-World Mobile Image Denoising Dataset with Efficient BaselinesCode2
Geodesic Diffusion Models for Medical Image-to-Image GenerationCode2
Restore-RWKV: Efficient and Effective Medical Image Restoration with RWKVCode2
CascadedGaze: Efficiency in Global Context Extraction for Image RestorationCode2
Fast-DDPM: Fast Denoising Diffusion Probabilistic Models for Medical Image-to-Image GenerationCode2
DualDn: Dual-domain Denoising via Differentiable ISPCode2
DnLUT: Ultra-Efficient Color Image Denoising via Channel-Aware Lookup TablesCode2
Dual-domain strip attention for image restorationCode2
Degradation-Aware Feature Perturbation for All-in-One Image RestorationCode2
Dynamic Pre-training: Towards Efficient and Scalable All-in-One Image RestorationCode2
Invisible Image Watermarks Are Provably Removable Using Generative AICode2
RenderDiffusion: Image Diffusion for 3D Reconstruction, Inpainting and GenerationCode2
Image Restoration with Mean-Reverting Stochastic Differential EquationsCode2
Make Explicit Calibration Implicit: Calibrate Denoiser Instead of the Noise ModelCode2
MaIR: A Locality- and Continuity-Preserving Mamba for Image RestorationCode2
Controlling Vision-Language Models for Multi-Task Image RestorationCode2
Ada-LISTA: Learned Solvers Adaptive to Varying ModelsCode1
Blind2Unblind: Self-Supervised Image Denoising with Visible Blind SpotsCode1
BiO-Net: Learning Recurrent Bi-directional Connections for Encoder-Decoder ArchitectureCode1
3D Quasi-Recurrent Neural Network for Hyperspectral Image DenoisingCode1
Deep Image PriorCode1
Deep Random Projector: Accelerated Deep Image PriorCode1
A cross Transformer for image denoisingCode1
Show:102550
← PrevPage 1 of 25Next →
All datasetsSIDDDNDELD SonyA7S2 x200SID SonyA7S2 x250ELD SonyA7S2 x100SID x100SID x300SID SonyA7S2 x100urban100 sigma15Urban100 sigma50BSD68 sigma50SID SonyA7S2 x300

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1CGNetPSNR (sRGB)40.39—Unverified
2KBNetPSNR (sRGB)40.35—Unverified
3NAFNetPSNR (sRGB)40.3—Unverified
4SSAMANPSNR (sRGB)40.08—Unverified
5PNGANPSNR (sRGB)40.07—Unverified
6RestormerPSNR (sRGB)40.02—Unverified
7HINetPSNR (sRGB)39.99—Unverified
8MAXIM-3SPSNR (sRGB)39.96—Unverified
9Uformer-BPSNR (sRGB)39.89—Unverified
10SRMNetPSNR (sRGB)39.72—Unverified
#ModelMetricClaimedVerifiedStatus
1DualDnPSNR (sRGB)40.59—Unverified
2PNGANPSNR (sRGB)40.18—Unverified
3SSAMANPSNR (sRGB)40.05—Unverified
4RestormerPSNR (sRGB)40.03—Unverified
5Uformer-BPSNR (sRGB)39.98—Unverified
6NBNetPSNR (sRGB)39.89—Unverified
7MIRNetPSNR (sRGB)39.88—Unverified
8MAXIM-3SPSNR (sRGB)39.84—Unverified
9MPRNetPSNR (sRGB)39.8—Unverified
10SADNetPSNR (sRGB)39.59—Unverified
#ModelMetricClaimedVerifiedStatus
1LLD*PSNR (Raw)44.95—Unverified
2PMNPSNR (Raw)44.51—Unverified
3SFRNPSNR (Raw)44.1—Unverified
4LLDPSNR (Raw)43.84—Unverified
5ELDPSNR (Raw)43.43—Unverified
6LRDPSNR (Raw)43.32—Unverified
7Paired Data(SID)PSNR (Raw)41.97—Unverified
8StarlightPSNR (Raw)40.86—Unverified
9ExposureDiffusion (UNet+ELD)PSNR (Raw)40.39—Unverified
10Noise FlowPSNR (Raw)39.23—Unverified
#ModelMetricClaimedVerifiedStatus
1LLD*PSNR (Raw)41.02—Unverified
2PMNPSNR (Raw)40.92—Unverified
3SFRNPSNR (Raw)40.22—Unverified
4LLDPSNR (Raw)39.76—Unverified
5Paired Data (SID)PSNR (Raw)39.6—Unverified
6ELDPSNR (Raw)39.44—Unverified
7LEDPSNR (Raw)39.34—Unverified
8LRDPSNR (Raw)39.25—Unverified
9StarlightPSNR (Raw)36.25—Unverified
10Noise FlowPSNR (Raw)35.8—Unverified
#ModelMetricClaimedVerifiedStatus
1LLD*PSNR (Raw)46.74—Unverified
2PMNPSNR (Raw)46.5—Unverified
3SFRNPSNR (Raw)46.02—Unverified
4LLDPSNR (Raw)45.61—Unverified
5ELDPSNR (Raw)45.45—Unverified
6LRDPSNR (Raw)44.95—Unverified
7Paired Data(SID)PSNR (Raw)44.47—Unverified
8StarlightPSNR (Raw)43.8—Unverified
9Noise FlowPSNR (Raw)41.05—Unverified
#ModelMetricClaimedVerifiedStatus
1LLD*PSNR (Raw)43.36—Unverified
2PMNPSNR (Raw)43.16—Unverified
3SFRNPSNR (Raw)42.29—Unverified
4LLDPSNR (Raw)42.1—Unverified
5SID (paired real data)PSNR (Raw)42.06—Unverified
6ELDPSNR (Raw)41.95—Unverified
7StarlightPSNR (Raw)40.47—Unverified
8Noise FlowPSNR (Raw)38.89—Unverified
#ModelMetricClaimedVerifiedStatus
1LLD*PSNR (Raw)37.8—Unverified
2PMNPSNR (Raw)37.77—Unverified
3SFRNPSNR (Raw)36.87—Unverified
4Paired Data(SID)PSNR (Raw)36.85—Unverified
5LLDPSNR (Raw)36.76—Unverified
6ELDPSNR (Raw)36.36—Unverified
7StarlightPSNR (Raw)32.99—Unverified
8Noise FlowPSNR (Raw)32.29—Unverified
#ModelMetricClaimedVerifiedStatus
1PMNPSNR (Raw)43.16—Unverified
2SFRNPSNR (Raw)42.29—Unverified
3LEDPSNR (Raw)41.98—Unverified
4ELDPSNR (Raw)41.95—Unverified
5LRDPSNR (Raw)41.95—Unverified
#ModelMetricClaimedVerifiedStatus
1AKDTAverage PSNR35.64—Unverified
2MaIR+PSNR35.42—Unverified
3MaIRPSNR35.35—Unverified
4SCUNet SCUNetAverage PSNR35.18—Unverified
#ModelMetricClaimedVerifiedStatus
1MaIR+PSNR30.41—Unverified
2MaIRPSNR30.3—Unverified
3SCUNet SCUNetPSNR30.14—Unverified
4AKDTPSNR29.82—Unverified
#ModelMetricClaimedVerifiedStatus
1MaIR+PSNR30.08—Unverified
2MaIRPSNR28.66—Unverified
#ModelMetricClaimedVerifiedStatus
1LEDPSNR (Raw)36.67—Unverified
2LRDPSNR (Raw)36.03—Unverified
#ModelMetricClaimedVerifiedStatus
1MaIR+PSNR33.3—Unverified
2MaIRPSNR33.22—Unverified
#ModelMetricClaimedVerifiedStatus
1R3LPSNR27.67—Unverified
#ModelMetricClaimedVerifiedStatus
1BRGMLPIPS0.24—Unverified
#ModelMetricClaimedVerifiedStatus
1BRGMLPIPS0.24—Unverified
#ModelMetricClaimedVerifiedStatus
1ExposureDiffusion (UNet+paired data)PSNR (Raw)36.82—Unverified
#ModelMetricClaimedVerifiedStatus
1PNGANPSNR40.78—Unverified
#ModelMetricClaimedVerifiedStatus
1PNGANPSNR40.55—Unverified
#ModelMetricClaimedVerifiedStatus
1absDLODRMSE0.07—Unverified