SOTAVerified

Deblurring

Deblurring is a computer vision task that involves removing the blurring artifacts from images or videos to restore the original, sharp content. Blurring can be caused by various factors such as camera shake, fast motion, and out-of-focus objects, and can result in a loss of detail and quality in the captured images. The goal of deblurring is to produce a clear, high-quality image that accurately represents the original scene.

( Image credit: Deblurring Face Images using Uncertainty Guided Multi-Stream Semantic Networks )

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 1–50 of 999 papers

TitleStatusHype
One Diffusion to Generate Them AllCode4
Blind Image Deblurring with Unknown Kernel Size and Substantial NoiseCode4
InstructIR: High-Quality Image Restoration Following Human InstructionsCode4
Zero-Shot Image Restoration Using Denoising Diffusion Null-Space ModelCode4
Simple Baselines for Image RestorationCode4
DocRes: A Generalist Model Toward Unifying Document Image Restoration TasksCode4
MedSegDiff: Medical Image Segmentation with Diffusion Probabilistic ModelCode3
Blind Image Restoration via Fast Diffusion InversionCode3
Deep learning in motion deblurring: current status, benchmarks and future prospectsCode3
BAD-Gaussians: Bundle Adjusted Deblur Gaussian SplattingCode3
Self-Rectifying Diffusion Sampling with Perturbed-Attention GuidanceCode3
MAXIM: Multi-Axis MLP for Image ProcessingCode3
Instruct-IPT: All-in-One Image Processing Transformer via Weight ModulationCode3
RobustSAM: Segment Anything Robustly on Degraded ImagesCode3
Rethinking Histology Slide Digitization Workflows for Low-Resource SettingsCode3
VRT: A Video Restoration TransformerCode3
AdaIR: Adaptive All-in-One Image Restoration via Frequency Mining and ModulationCode3
DarkIR: Robust Low-Light Image RestorationCode3
AdaRevD: Adaptive Patch Exiting Reversible Decoder Pushes the Limit of Image DeblurringCode3
Event-Enhanced Blurry Video Super-ResolutionCode3
MaIR: A Locality- and Continuity-Preserving Mamba for Image RestorationCode2
LoFormer: Local Frequency Transformer for Image DeblurringCode2
MB-TaylorFormer V2: Improved Multi-branch Linear Transformer Expanded by Taylor Formula for Image RestorationCode2
Deblur4DGS: 4D Gaussian Splatting from Blurry Monocular VideoCode2
KBNet: Kernel Basis Network for Image RestorationCode2
Image Restoration with Mean-Reverting Stochastic Differential EquationsCode2
Blur-aware Spatio-temporal Sparse Transformer for Video DeblurringCode2
LEDNet: Joint Low-light Enhancement and Deblurring in the DarkCode2
Bokehlicious: Photorealistic Bokeh Rendering with Controllable AperturesCode2
Blur Interpolation Transformer for Real-World Motion from BlurCode2
I^2SB: Image-to-Image Schrödinger BridgeCode2
Learning Truncated Causal History Model for Video RestorationCode2
Flow Priors for Linear Inverse Problems via Iterative Corrupted Trajectory MatchingCode2
CRNet: A Detail-Preserving Network for Unified Image Restoration and Enhancement TaskCode2
Fourier Priors-Guided Diffusion for Zero-Shot Joint Low-Light Enhancement and DeblurringCode2
Efficient Frequency Domain-based Transformers for High-Quality Image DeblurringCode2
Correlation Matching Transformation Transformers for UHD Image RestorationCode2
BAD-NeRF: Bundle Adjusted Deblur Neural Radiance FieldsCode2
EAMamba: Efficient All-Around Vision State Space Model for Image RestorationCode2
Efficient Visual State Space Model for Image DeblurringCode2
Beyond Text: Frozen Large Language Models in Visual Signal ComprehensionCode2
Hierarchical Integration Diffusion Model for Realistic Image DeblurringCode2
Diffusion Prior-Based Amortized Variational Inference for Noisy Inverse ProblemsCode2
EnsIR: An Ensemble Algorithm for Image Restoration via Gaussian Mixture ModelsCode2
DocDiff: Document Enhancement via Residual Diffusion ModelsCode2
Event-guided Multi-patch Network with Self-supervision for Non-uniform Motion DeblurringCode2
Diffusion Models for Image Restoration and Enhancement -- A Comprehensive SurveyCode2
Diffusion Posterior Sampling for General Noisy Inverse ProblemsCode2
Denoising as Adaptation: Noise-Space Domain Adaptation for Image RestorationCode2
Exposure Bracketing Is All You Need For A High-Quality ImageCode2
Show:102550
← PrevPage 1 of 20Next →
All datasetsGoProHIDE (trained on GOPRO)RealBlur-R (trained on GoPro)RealBlur-JRealBlur-RRealBlur-J (trained on GoPro)RSBlurMSU BASEDDVDBeam-Splitter Deblurring (BSD)BASEDREDS

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1BSSTNetPSNR35.98—Unverified
2GShift-NetPSNR35.88—Unverified
3DLEFNetPSNR35.61—Unverified
4EFNetPSNR35.46—Unverified
5NIREPSNR35.03—Unverified
6VRTPSNR34.81—Unverified
7AdaRevDPSNR34.6—Unverified
8TurtlePSNR34.5—Unverified
9ID-Blau (FFTformer)PSNR34.36—Unverified
10XYScanNetPSNR33.91—Unverified
#ModelMetricClaimedVerifiedStatus
1MAXIMPSNR (sRGB)32.83—Unverified
2AdaRevDPSNR (sRGB)32.35—Unverified
3ID-Blau (FFTformer)PSNR (sRGB)31.94—Unverified
4LoFormerPSNR (sRGB)31.86—Unverified
5CAPTNetPSNR (sRGB)31.86—Unverified
6XYScanNetPSNR (sRGB)31.74—Unverified
7DeepRFT+PSNR (sRGB)31.66—Unverified
8ID-Blau (Restormer)PSNR (sRGB)31.66—Unverified
9FFTformerPSNR (sRGB)31.62—Unverified
10LaKDNetPSNR (sRGB)31.58—Unverified
#ModelMetricClaimedVerifiedStatus
1AdaRevDPSNR (sRGB)36.53—Unverified
2ALGNetPSNR (sRGB)36.35—Unverified
3Uformer-BPSNR (sRGB)36.22—Unverified
4RestormerPSNR (sRGB)36.19—Unverified
5DeepRFTPSNR (sRGB)36.11—Unverified
6DeblurDiNAT-LPSNR (sRGB)36.09—Unverified
7LaKDNetPSNR (sRGB)36.08—Unverified
8MPRNetPSNR (sRGB)35.99—Unverified
9MSSNetPSNR (sRGB)35.93—Unverified
10MAXIMPSNR (sRGB)35.78—Unverified
#ModelMetricClaimedVerifiedStatus
1AdaRevDPSNR (sRGB)33.96—Unverified
2MLWNetPSNR (sRGB)33.84—Unverified
3ID-Blau (Stripformer)PSNR (sRGB)33.77—Unverified
4ID-Blau (Restormer)PSNR (sRGB)33.11—Unverified
5ALGNetPSNR (sRGB)32.94—Unverified
6LoFormerPSNR (sRGB)32.9—Unverified
7ID-Blau (FFTformer)PSNR (sRGB)32.88—Unverified
8MAXIMPSNR (sRGB)32.84—Unverified
9DeepRFT+PSNR (sRGB)32.63—Unverified
10FFTformerPSNR (sRGB)32.62—Unverified
#ModelMetricClaimedVerifiedStatus
1AdaRevDPSNR (sRGB)41.19—Unverified
2ALGNetPSNR (sRGB)41.16—Unverified
3ID-Blau (Stripformer)PSNR (sRGB)41.06—Unverified
4MLWNetPSNR (sRGB)40.69—Unverified
5ID-Blau (FFTformer)PSNR (sRGB)40.45—Unverified
6ID-Blau (Restormer)PSNR (sRGB)40.31—Unverified
7LoFormerPSNR (sRGB)40.23—Unverified
8FFTformerPSNR (sRGB)40.11—Unverified
9DeepRFT+PSNR (sRGB)40.01—Unverified
10StripformerPSNR (sRGB)39.84—Unverified
#ModelMetricClaimedVerifiedStatus
1AdaRevDPSNR (sRGB)30.12—Unverified
2UFPNetPSNR (sRGB)29.87—Unverified
3ALGNetPSNR (sRGB)29.12—Unverified
4Uformer-BPSNR (sRGB)29.06—Unverified
5DeblurDiNAT-LPSNR (sRGB)28.98—Unverified
6RestormerPSNR (sRGB)28.96—Unverified
7DeepRFT+PSNR (sRGB)28.88—Unverified
8MAXIMPSNR (sRGB)28.83—Unverified
9MSSNetPSNR (sRGB)28.79—Unverified
10MPRNetPSNR (sRGB)28.7—Unverified
#ModelMetricClaimedVerifiedStatus
1MLWNetAverage PSNR34.94—Unverified
2SegDeblurAverage PSNR34.63—Unverified
3SFNetAverage PSNR34.35—Unverified
4FSNetAverage PSNR34.31—Unverified
5IRNextAverage PSNR34.08—Unverified
6ConvIRAverage PSNR34.06—Unverified
7Uformer-BAverage PSNR33.98—Unverified
8RestormerAverage PSNR33.69—Unverified
9MPRNetAverage PSNR33.61—Unverified
10MIMO-UNet+Average PSNR33.37—Unverified
#ModelMetricClaimedVerifiedStatus
1NAFNet (REDS)Subjective2.84—Unverified
2VRT (GoPro)Subjective2.39—Unverified
3VRT (REDS)Subjective1.57—Unverified
4DeblurGAN InceptionSubjective1.04—Unverified
5MAXIM (REDS)Subjective1.01—Unverified
6Deeprft (GoPro)Subjective0.54—Unverified
7Deeprft (REDS)Subjective0.46—Unverified
8MPR localSubjective0.44—Unverified
9MAXIM (GoPro)Subjective0.21—Unverified
10Restormer localSubjective0.12—Unverified
#ModelMetricClaimedVerifiedStatus
1BSSTNetPSNR34.95—Unverified
2RVRTPSNR34.92—Unverified
3GShift-NetPSNR34.69—Unverified
4VRTPSNR34.27—Unverified
5FGSTPSNR33.5—Unverified
6STDANPSNR33.05—Unverified
7FGSTPSNR33.03—Unverified
8MB2DPSNR32.34—Unverified
9CDVD-TSPPSNR32.13—Unverified
10STFANPSNR31.27—Unverified
#ModelMetricClaimedVerifiedStatus
1TurtlePSNR33.58—Unverified
2CDVD-TSPPSNR31.58—Unverified
3ESTRNNPSNR31.39—Unverified
4PVDNetPSNR31.35—Unverified
5DBNPSNR31.21—Unverified
#ModelMetricClaimedVerifiedStatus
1VRT (GoPro)PSNR31.43—Unverified
2DeblurGAN InceptionPSNR31.17—Unverified
3VRT (REDS)PSNR30.98—Unverified
4MAXIM (REDS)PSNR30.66—Unverified
#ModelMetricClaimedVerifiedStatus
1VRTAverage PSNR36.79—Unverified
2EDVR_DeblurAverage PSNR34.8—Unverified
3DeblurGANAverage PSNR24.09—Unverified
#ModelMetricClaimedVerifiedStatus
1MIMO-UNet + Realistic blurAverage PSNR32.08—Unverified
2SRN-Deblur + Realistic blurAverage PSNR32.06—Unverified
#ModelMetricClaimedVerifiedStatus
1110 Images, 4*4 Stitching, Exact Accuracy2—Unverified
#ModelMetricClaimedVerifiedStatus
1MAXIM-3SPSNR32.83—Unverified
#ModelMetricClaimedVerifiedStatus
1OursMAE0.04—Unverified