Restormer: Efficient Transformer for High-Resolution Image Restoration

2021/11/18 by Syed Waqas Zamir, Aditya Arora, Zamir, Syed Waqas +10 · 351 citations
Computer Science · Engineering · #Advanced Image Processing Techniques #Artificial intelligence #Computer Vision and Pattern Recognition (cs.CV) #Computer science #Computer vision #Deblurring #FOS: Computer and information sciences #Image (mathematics) #Image Processing Techniques and Applications #Image and Signal Denoising Methods #Image processing #Image restoration #Image warping #Inpainting #Pattern recognition (psychology) #Pixel #Transformer #cs.CV

paper · pdf · doi:10.48550/arxiv.2111.09881

published in arXiv (Cornell University) (Cornell University) · Accepted at CVPR 2022. #CVPR2022

openalex publication_date 2021/11/18 · arxiv created 2022/03/11 · arxiv updated 2022/03/14 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/04

Abstract

Since convolutional neural networks (CNNs) perform well at learning generalizable image priors from large-scale data, these models have been extensively applied to image restoration and related tasks. Recently, another class of neural architectures, Transformers, have shown significant performance gains on natural language and high-level vision tasks. While the Transformer model mitigates the shortcomings of CNNs (i.e., limited receptive field and inadaptability to input content), its computational complexity grows quadratically with the spatial resolution, therefore making it infeasible to apply to most image restoration tasks involving high-resolution images. In this work, we propose an efficient Transformer model by making several key designs in the building blocks (multi-head attention and feed-forward network) such that it can capture long-range pixel interactions, while still remaining applicable to large images. Our model, named Restoration Transformer (Restormer), achieves state-of-the-art results on several image restoration tasks, including image deraining, single-image motion deblurring, defocus deblurring (single-image and dual-pixel data), and image denoising (Gaussian grayscale/color denoising, and real image denoising). The source code and pre-trained models are available at https://github.com/swz30/Restormer.

Citations

• Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising
• Self-Attention Generative Adversarial Networks
• Residual Non-local Attention Networks for Image Restoration
• Language Models are Few-Shot Learners
• Deformable DETR: Deformable Transformers for End-to-End Object Detection
• Tokens-to-Token ViT: Training Vision Transformers from Scratch on ImageNet
• Swin Transformer: Hierarchical Vision Transformer using Shifted Windows
• Uformer: A General U-Shaped Transformer for Image Restoration
• Multi-Stage Progressive Image Restoration
• Rethinking Semantic Segmentation from a Sequence-to-Sequence Perspective with Transformers
• Pre-Trained Image Processing Transformer
• Pyramid Vision Transformer: A Versatile Backbone for Dense Prediction without Convolutions
• SwinIR: Image Restoration Using Swin Transformer

Cited by

• Image Denoising Using Global and Local Circulant Representation
• Stream-DiffVSR: Low-Latency Streamable Video Super-Resolution via Auto-Regressive Diffusion
• Stabilizing Deep Reconstruction Operators with Contractive Anchoring
• Low-light Image Enhancement via Multi-scale Attention combined with Fourier Transform
• RPG-VST: Robust Poisson-Gaussian Variance Stabilization for Blind RAW Denoising
• Zhinv: Real-time hub-height wind field reconstruction using only local sparse observations
• Hyperspectral Intrinsic Decomposition: Joint Recovery of Reflectance and Photometric Components for Non-Lambertian Scenes
• Noise-Free One-Step LoRA for Task-Driven Image Restoration with Diffusion Priors
• Frequency-Aware Dual-Stream Learning for Balanced Realism and Fidelity in Electron Microscopy Imaging
• Small, Bias-Free, Blind and Convolutional Denoiser: A compact ConvNeXt U-Net for blind Gaussian color-image denoising
• Multinex: Lightweight Low-light Image Enhancement via Multi-prior Retinex
• SDUM: A Scalable Deep Unrolled Model for Universal MRI Reconstruction
• Generative Multi-Focus Image Fusion
• Multi-Grained Text-Guided Image Fusion for Multi-Exposure and Multi-Focus Scenarios
• Degradation-Aware Metric Prompting for Hyperspectral Image Restoration
• JDPNet: A Network Based on Joint Degradation Processing for Underwater Image Enhancement
• Generative Latent Coding for Ultra-Low Bitrate Image Compression
• Learning to Refocus with Video Diffusion Models
• Active Convolved Illumination with Deep Transfer Learning for Complex Beam Transmission through Atmospheric Turbulence
• Mamba-Based Modality Disentanglement Network for Multi-Contrast MRI Reconstruction
• SplatBright: Generalizable Low-Light Scene Reconstruction from Sparse Views via Physically-Guided Gaussian Enhancement
• VizDefender: Unmasking Visualization Tampering through Proactive Localization and Intent Inference
• Restore-R1: Efficient Image Restoration Agents via Reinforcement Learning with Multimodal LLM Perceptual Feedback
• Multi-level distortion-aware deformable network for omnidirectional image super-resolution
• Generative Refocusing: Flexible Defocus Control from a Single Image
• SLCFormer: Spectral-Local Context Transformer with Physics-Grounded Flare Synthesis for Nighttime Flare Removal
• MVGSR: Multi-View Consistent 3D Gaussian Super-Resolution via Epipolar Guidance
• TAT: Task-Adaptive Transformer for All-in-One Medical Image Restoration
• Bridging Fidelity-Reality with Controllable One-Step Diffusion for Image Super-Resolution
• Physics-Informed Video Flare Synthesis and Removal Leveraging Motion Independence between Flare and Scene
• Generative Manifold Distillation: Aligning Restoration Trajectories with Natural Image Prior
• ClusIR: Towards Cluster-Guided All-in-One Image Restoration
• Unleashing Degradation-Carrying Features in Symmetric U-Net: Simpler and Stronger Baselines for All-in-One Image Restoration
• Content-Adaptive Image Retouching Guided by Attribute-Based Text Representation
• FoundIR-v2: Optimizing Pre-Training Data Mixtures for Image Restoration Foundation Model
• CARD: Correlation Aware Restoration with Diffusion
• Fourier-RWKV: A Multi-State Perception Network for Efficient Image Dehazing
• DGGAN: Degradation Guided Generative Adversarial Network for Real-time Endoscopic Video Enhancement
• UARE: A Unified Vision-Language Model for Image Quality Assessment, Restoration, and Enhancement
• Experts-Guided Unbalanced Optimal Transport for ISP Learning from Unpaired and/or Paired Data
• Consist-Retinex: One-Step Noise-Emphasized Consistency Training Accelerates High-Quality Retinex Enhancement
• Physics-Informed Graph Neural Network with Frequency-Aware Learning for Optical Aberration Correction
• EvoIR: Towards All-in-One Image Restoration via Evolutionary Frequency Modulation
• UnwrapDiff: Conditional Diffusion for Robust InSAR Phase Unwrapping
• FMA-Net++: Motion- and Exposure-Aware Real-World Joint Video Super-Resolution and Deblurring
• Beyond the Ground Truth: Enhanced Supervision for Image Restoration
• BlurDM: A Blur Diffusion Model for Image Deblurring
• Traffic Image Restoration under Adverse Weather via Frequency-Aware Mamba
• Does Head Pose Correction Improve Biometric Facial Recognition?
• 2-Shots in the Dark: Low-Light Denoising with Minimal Data Acquisition
• Progressive Image Restoration via Text-Conditioned Video Generation
• Exploring the Potentials of Spiking Neural Networks for Image Deraining
• TPCNet: Triple physical constraints for Low-light Image Enhancement
• World Model Robustness via Surprise Recognition
• PAGen: Phase-guided Amplitude Generation for Domain-adaptive Object Detection
• IRPO: Boosting Image Restoration via Post-training GRPO
• HIMOSA: Efficient Remote Sensing Image Super-Resolution with Hierarchical Mixture of Sparse Attention
• PG-ControlNet: A Physics-Guided ControlNet for Generative Spatially Varying Image Deblurring
• DeepRFTv2: Kernel-level Learning for Image Deblurring
• MODEST: Multi-Optics Depth-of-Field Stereo Dataset
• VICoT-Agent: A Vision-Interleaved Chain-of-Thought Framework for Interpretable Multimodal Reasoning and Scalable Remote Sensing Analysis
• Test-Time Preference Optimization for Image Restoration
• ProvRain: Rain-Adaptive Denoising and Vehicle Detection via MobileNet-UNet and Faster R-CNN
• NAF: Zero-Shot Feature Upsampling via Neighborhood Attention Filtering
• SD-PSFNet: Sequential and Dynamic Point Spread Function Network for Image Deraining
• VLM-Augmented Degradation Modeling for Image Restoration Under Adverse Weather Conditions
• Evaluating Low-Light Image Enhancement Across Multiple Intensity Levels
• CompEvent: Complex-valued Event-RGB Fusion for Low-light Video Enhancement and Deblurring
• Semantics and Content Matter: Towards Multi-Prior Hierarchical Mamba for Image Deraining
• Enhancing Diffusion-based Restoration Models via Difficulty-Adaptive Reinforcement Learning with IQA Reward
• A lightweight progressive aggregation network for multi-contrast MRI super-resolution
• DINO-Detect: A Simple yet Effective Framework for Blur-Robust AI-Generated Image Detection
• DCA-LUT: Deep Chromatic Alignment with 5D LUT for Purple Fringing Removal
• Text-Guided Channel Perturbation and Pretrained Knowledge Integration for Unified Multi-Modality Image Fusion
• From Events to Clarity: The Event-Guided Diffusion Framework for Dehazing
• Physically Interpretable Multi-Degradation Image Restoration via Deep Unfolding and Explainable Convolution
• Equivariant Sampling for Improving Diffusion Model-based Image Restoration
• Hierarchical Spatial-Frequency Aggregation for Spectral Deconvolution Imaging
• Spatial-Frequency Enhanced Mamba for Multi-Modal Image Fusion
• Hybrid CNN-ViT Framework for Motion-Blurred Scene Text Restoration
• HarmoQ: Harmonized Post-Training Quantization for High-Fidelity Image
• UHDRes: Ultra-High-Definition Image Restoration via Dual-Domain Decoupled Spectral Modulation
• Learning to Restore Multi-Degraded Images via Ingredient Decoupling and Task-Aware Path Adaptation
• Diffusion Transformer meets Multi-level Wavelet Spectrum for Single Image Super-Resolution
• Learnable Fractional Reaction-Diffusion Dynamics for Under-Display ToF Imaging and Beyond
• VisionCAD: An Integration-Free Radiology Copilot Framework
• Trans-defense: Transformer-based Denoiser for Adversarial Defense with Spatial-Frequency Domain Representation
• CASIAL: Geometric Distortion Robust Image Watermarking
• Larger Hausdorff Dimension in Scanning Pattern Facilitates Mamba-Based Methods in Low-Light Image Enhancement
• HYDRA: HYbrid knowledge Distillation and spectral Reconstruction Algorithm for high channel hyperspectral camera applications
• Residual Diffusion Bridge Model for Image Restoration
• Low-Light Image Enhancement Using Gamma Learning And Attention-Enabled Encoder-Decoder Networks
• UniMedVL: Unifying Medical Multimodal Understanding And Generation Through Observation-Knowledge-Analysis
• RaindropGS: A Benchmark for 3D Gaussian Splatting under Raindrop Conditions
• CharDiff-LP: A Diffusion Model with Character-Level Guidance for License Plate Image Restoration
• Boosting Fidelity for Pre-Trained-Diffusion-Based Low-Light Image Enhancement via Condition Refinement
• Rethinking Nighttime Image Deraining via Learnable Color Space Transformation
• SDPA++: A General Framework for Self-Supervised Denoising with Patch Aggregation
• WaMaIR: Image Restoration via Multiscale Wavelet Convolutions and Mamba-based Channel Modeling with Texture Enhancement
• Leveraging Learned Image Prior for 3D Gaussian Compression
• Pruning Overparameterized Multi-Task Networks for Degraded Web Image Restoration
• NTIRE 2025 Challenge on Low Light Image Enhancement: Methods and Results
• Universal Image Restoration Pre-training via Masked Degradation Classification
• Efficient Real-World Deblurring using Single Images: AIM 2025 Challenge Report
• Zero-Shot CFC: Fast Real-World Image Denoising based on Cross-Frequency Consistency
• AngularFuse: A Closer Look at Angle-based Perception for Spatial-Sensitive Multi-Modality Image Fusion
• Dynamic Gaussian Splatting from Defocused and Motion-blurred Monocular Videos
• Enabling High-Quality In-the-Wild Imaging from Severely Aberrated Metalens Bursts
• BurstDeflicker: A Benchmark Dataset for Flicker Removal in Dynamic Scenes
• FlareX: A Physics-Informed Dataset for Lens Flare Removal via 2D Synthesis and 3D Rendering
• Clear Roads, Clear Vision: Advancements in Multi-Weather Restoration for Smart Transportation
• Enhancing Infrared Vision: Progressive Prompt Fusion Network and Benchmark
• LinearSR: Unlocking Linear Attention for Stable and Efficient Image Super-Resolution
• Latent Harmony: Synergistic Unified UHD Image Restoration via Latent Space Regularization and Controllable Refinement
• FideDiff: Efficient Diffusion Model for High-Fidelity Image Motion Deblurring
• Joint Deblurring and 3D Reconstruction for Macrophotography
• NPN: Non-Linear Projections of the Null-Space for Imaging Inverse Problems
• Transforming Noise Distributions with Histogram Matching: Towards a Single Denoiser for All
• DeRainMamba: A Frequency-Aware State Space Model with Detail Enhancement for Image Deraining
• AIM 2025 Challenge on Real-World RAW Image Denoising
• TDiff: Thermal Plug-And-Play Prior with Patch-Based Diffusion
• Diffusion Models for Low-Light Image Enhancement: A Multi-Perspective Taxonomy and Performance Analysis
• PocketSR: The Super-Resolution Expert in Your Pocket Mobiles
• Net2Net: When Un-trained Meets Pre-trained Networks for Robust Real-World Denoising
• Extreme Blind Image Restoration via Prompt-Conditioned Information Bottleneck
• DiffCamera: Arbitrary Refocusing on Images
• PRISM: Progressive Rain removal with Integrated State-space Modeling
• FlowLUT: Efficient Image Enhancement via Differentiable LUTs and Iterative Flow Matching
• WeatherCycle: Unpaired Multi-Weather Restoration via Color Space Decoupled Cycle Learning
• LucidFlux: Caption-Free Universal Image Restoration via a Large-Scale Diffusion Transformer
• Unleashing the Potential of the Semantic Latent Space in Diffusion Models for Image Dehazing
• CoRE-UIR: Prior-guided common and residual experts for efficient all-in-one remote sensing image restoration
• ISALux: Illumination and Segmentation Aware Transformer Employing Mixture of Experts for Low Light Image Enhancement
• CATformer: Contrastive Adversarial Transformer for Image Super-Resolution
• Degradation-Aware All-in-One Image Restoration via Latent Prior Encoding
• PRNU-Bench: A Novel Benchmark and Model for PRNU-Based Camera Identification
• SmokeSeer: 3D Gaussian Splatting for Smoke Removal and Scene Reconstruction
• \mathttM³VIR: A Large-Scale Multi-Modality Multi-View Synthesized Benchmark Dataset for Image Restoration and Content Creation
• CGTGait: Collaborative Graph and Transformer for Gait Emotion Recognition
• FoBa: A Foreground-Background co-Guided Method and New Benchmark for Remote Sensing Semantic Change Detection
• QWD-GAN: Quality-aware Wavelet-driven GAN for Unsupervised Medical Microscopy Images Denoising
• Not All Degradations Are Equal: A Targeted Feature Denoising Framework for Generalizable Image Super-Resolution
• NDLPNet: A Location-Aware Nighttime Deraining Network and a Real-World Benchmark Dataset
• AIM 2025 Low-light RAW Video Denoising Challenge: Dataset, Methods and Results
• Cross-Distribution Diffusion Priors-Driven Iterative Reconstruction for Sparse-View CT
• FoundDiff: Foundational Diffusion Model for Generalizable Low-Dose CT Denoising
• SAGA: Selective Adaptive Gating for Efficient and Expressive Linear Attention
• SmokeBench: A Real-World Dataset for Surveillance Image Desmoking in Early-Stage Fire Scenes
• Using KL-Divergence to Focus Frequency Information in Low-Light Image Enhancement
• Generative AI for Multimedia Communication: Recent Advances, An Information-Theoretic Framework, and Future Opportunities
• RAM++: Robust Representation Learning via Adaptive Mask for All-in-One Image Restoration
• Unrolling Graph-based Douglas-Rachford Algorithm for Image Interpolation with Informed Initialization
• Impact of a Sharpness Based Loss Function for Removing Out-of-Focus Blur
• WeatherBench: A Real-World Benchmark Dataset for All-in-One Adverse Weather Image Restoration
• Geometric Analysis of Magnetic Labyrinthine Stripe Evolution via Deep Learning Segmentation
• USCTNet: A deep unfolding nuclear-norm optimization solver for physically consistent HSI reconstruction
• A novel method and dataset for depth-guided image deblurring from smartphone Lidar
• BIR-Adapter: A Low-Complexity Diffusion Model Adapter for Blind Image Restoration
• AIM 2025 Challenge on High FPS Motion Deblurring: Methods and Results
• Stabilizing RED using the Koopman Operator
• RED: Robust Event-Guided Motion Deblurring with Modality-Specific Disentanglement
• WIPUNet: A Physics-inspired Network with Weighted Inductive Biases for Image Denoising
• Solving Imaging Inverse Problems Using Plug-and-Play Denoisers: Regularization and Optimization Perspectives
• DroneSR: Rethinking Few-shot Thermal Image Super-Resolution from Drone-based Perspective
• MILO: A Lightweight Perceptual Quality Metric for Image and Latent-Space Optimization
• DarkVRAI: Capture-Condition Conditioning and Burst-Order Selective Scan for Low-light RAW Video Denoising
• IDF: Iterative Dynamic Filtering Networks for Generalizable Image Denoising
• MoCHA-former: Moiré-Conditioned Hybrid Adaptive Transformer for Video Demoiréing
• GeMS: Efficient Gaussian Splatting for Extreme Motion Blur
• DIME-Net: A Dual-Illumination Adaptive Enhancement Network Based on Retinex and Mixture-of-Experts
• Learning to See Through Flare
• MF-LPR²: Multi-Frame License Plate Image Restoration and Recognition using Optical Flow
• Frequency-Driven Inverse Kernel Prediction for Single Image Defocus Deblurring
• Hybrid Deep Reconstruction for Vignetting-Free Upconversion Imaging through Scattering in ENZ Materials
• SNNSIR: A Simple Spiking Neural Network for Stereo Image Restoration
• WXSOD: A Benchmark for Robust Salient Object Detection in Adverse Weather Conditions
• MBMamba: When Memory Buffer Meets Mamba for Structure-Aware Image Deblurring
• EvTurb: Event Camera Guided Turbulence Removal
• Efficient Image Denoising Using Global and Local Circulant Representation
• Towards Perfection: Building Inter-component Mutual Correction for Retinex-based Low-light Image Enhancement
• SelfHVD: Self-Supervised Handheld Video Deblurring
• TAP: Parameter-efficient Task-Aware Prompting for Adverse Weather Removal
• CMAMRNet: A Contextual Mask-Aware Network Enhancing Mural Restoration Through Comprehensive Mask Guidance
• Towards Unified Image Deblurring using a Mixture-of-Experts Decoder
• ETA: Energy-based Test-time Adaptation for Depth Completion
• Lightweight Quad Bayer HybridEVS Demosaicing via State Space Augmented Cross-Attention
• AU-IQA: A Benchmark Dataset for Perceptual Quality Assessment of AI-Enhanced User-Generated Content
• Extending Foundational Monocular Depth Estimators to Fisheye Cameras with Calibration Tokens
• RetinexDual: Retinex-based Dual Nature Approach for Generalized Ultra-High-Definition Image Restoration
• Uncertainty-Aware Spatial Color Correlation for Low-Light Image Enhancement
• Bridging Diffusion Models and 3D Representations: A 3D Consistent Super-Resolution Framework
• MetaScope: Optics-Driven Neural Network for Ultra-Micro Metalens Endoscopy
• Beyond Illumination: Fine-Grained Detail Preservation in Extreme Dark Image Restoration
• Towards Robust Image Denoising with Scale Equivariance
• After the Party: Navigating the Mapping From Color to Ambient Lighting
• DeflareMamba: Hierarchical Vision Mamba for Contextually Consistent Lens Flare Removal
• Viscosity Stabilized Plug-and-Play Reconstruction
• Exploring Fourier Prior and Event Collaboration for Low-Light Image Enhancement
• Latent Diffusion Based Face Enhancement under Degraded Conditions for Forensic Face Recognition
• UniLDiff: Unlocking the Power of Diffusion Priors for All-in-One Image Restoration
• MRpro - open PyTorch-based MR reconstruction and processing package
• Robust Adverse Weather Removal via Spectral-based Spatial Grouping
• Exploiting Diffusion Prior for Task-driven Image Restoration
• From Sharp to Blur: Unsupervised Domain Adaptation for 2D Human Pose Estimation Under Extreme Motion Blur Using Event Cameras
• Moiré Zero: An Efficient and High-Performance Neural Architecture for Moiré Removal
• SAIGFormer: A Spatially-Adaptive Illumination-Guided Network for Low-Light Image Enhancement
• Neural network enabled wide field-of-view imaging with hyperbolic metalenses
• ModalFormer: Multimodal Transformer for Low-Light Image Enhancement
• GT-Mean Loss: A Simple Yet Effective Solution for Brightness Mismatch in Low-Light Image Enhancement
• MoFRR: Mixture of Diffusion Models for Face Retouching Restoration
• Event-Based De-Snowing for Autonomous Driving
• Continual Learning-Based Unified Model for Unpaired Image Restoration Tasks
• BokehDiff: Neural Lens Blur with One-Step Diffusion
• Degradation-Consistent Learning via Bidirectional Diffusion for Low-Light Image Enhancement
• DiNAT-IR: Exploring Dilated Neighborhood Attention for High-Quality Image Restoration
• Grounding Degradations in Natural Language for All-In-One Video Restoration
• Exploring Scalable Unified Modeling for General Low-Level Vision
• PolarAnything: Diffusion-based Polarimetric Image Synthesis
• Learning Deblurring Texture Prior from Unpaired Data with Diffusion Model
• Global Modeling Matters: A Fast, Lightweight and Effective Baseline for Efficient Image Restoration
• Efficient Dual-domain Image Dehazing with Haze Prior Perception
• Toward Robust and 3D-Aware RGB-NIR Imaging in the Dark
• Generative Latent Kernel Modeling for Blind Motion Deblurring
• Generalizable 7T T1-map Synthesis from 1.5T and 3T T1 MRI with an Efficient Transformer Model
• Degradation-Agnostic Statistical Facial Feature Transformation for Blind Face Restoration in Adverse Weather Conditions
• Motion-Aware Adaptive Pixel Pruning for Efficient Local Motion Deblurring
• Diffusion-Assisted Frequency Attention Model for Whole-body Low-field MRI Reconstruction
• DFYP: A Dynamic Fusion Framework with Spectral Channel Attention and Adaptive Operator learning for Crop Yield Prediction
• Reviving Cultural Heritage: A Novel Approach for Comprehensive Historical Document Restoration
• Towards Spatially-Varying Gain and Binning
• Low-Light Enhancement via Encoder-Decoder Network with Illumination Guidance
• PixIE: Prompted Pixel-Space Low-Light Image Enhancement
• Modulate and Reconstruct: Learning Hyperspectral Imaging from Misaligned Smartphone Views
• Enhancing Multi-Exposure High Dynamic Range Imaging with Overlapped Codebook for Improved Representation Learning
• Towards Controllable Real Image Denoising with Camera Parameters
• Oneta: Multi-Style Image Enhancement Using Eigentransformation Functions
• HazeMatching: Dehazing Light Microscopy Images with Guided Conditional Flow Matching
• Design and Evaluation of Deep Learning-Based Dual-Spectrum Image Fusion Methods
• EAMamba: Efficient All-Around Vision State Space Model for Image Restoration
• Physical Degradation Model-Guided Interferometric Hyperspectral Reconstruction with Unfolding Transformer
• Elucidating and Endowing the Diffusion Training Paradigm for General Image Restoration
• 3D Scene-Camera Representation with Joint Camera Photometric Optimization
• A Comparative Study of NAFNet Baselines for Image Restoration
• M2Restore: Mixture-of-Experts-based Mamba-CNN Fusion Framework for All-in-One Image Restoration
• Enhancing Image Restoration Transformer via Adaptive Translation Equivariance
• Frequency-Domain Fusion Transformer for Image Inpainting
• A Multi-Scale Spatial Attention-Based Zero-Shot Learning Framework for Low-Light Image Enhancement
• Reversing Flow for Image Restoration
• Visual-Instructed Degradation Diffusion for All-in-One Image Restoration
• Learning Multi-scale Spatial-frequency Features for Image Denoising
• Demystifying the Visual Quality Paradox in Multimodal Large Language Models
• NTIRE 2025 Image Shadow Removal Challenge Report
• Unsupervised Imaging Inverse Problems with Diffusion Distribution Matching
• Zero-shot Denoising via Neural Compression: Theoretical and algorithmic framework
• SPC to 3D: Novel View Synthesis from Binary SPC via I2I translation
• A Preliminary Study on GPT-Image Generation Model for Image Restoration
• Bidirectional Image-Event Guided Fusion Framework for Low-Light Image Enhancement
• SSH-Net: A Self-Supervised and Hybrid Network for Noisy Image Watermark Removal
• UniRes: Universal Image Restoration for Complex Degradations
• F2T2-HiT: A U-Shaped FFT Transformer and Hierarchical Transformer for Reflection Removal
• FEAT: Full-Dimensional Efficient Attention Transformer for Medical Video Generation
• Image Restoration via Multi-domain Learning
• ControlMambaIR: Conditional Controls with State-Space Model for Image Restoration
• Towards Better De-raining Generalization via Rainy Characteristics Memorization and Replay
• CLIP-driven rain perception: Adaptive deraining with pattern-aware network routing and mask-guided cross-attention
• Region-of-Interest-Guided Deep Joint Source-Channel Coding for Image Transmission
• NTIRE 2025 Challenge on RAW Image Restoration and Super-Resolution
• AceVFI: A Comprehensive Survey of Advances in Video Frame Interpolation
• Optimal Weighted Convolution for Classification and Denosing
• Boosting All-in-One Image Restoration via Self-Improved Privilege Learning
• Efficient RAW Image Deblurring with Adaptive Frequency Modulation
• Proximal Algorithm Unrolling: Flexible and Efficient Reconstruction Networks for Single-Pixel Imaging
• iHDR: Iterative HDR Imaging with Arbitrary Number of Exposures
• URWKV: Unified RWKV Model with Multi-state Perspective for Low-light Image Restoration
• Learnable Burst-Encodable Time-of-Flight Imaging for High-Fidelity Long-Distance Depth Sensing
• From Controlled Scenarios to Real-World: Cross-Domain Degradation Pattern Matching for All-in-One Image Restoration
• Multi-View Learning with Context-Guided Receptance for Image Denoising
• BaryIR: Learning Multi-Source Unified Representation in Continuous Barycenter Space for Generalizable All-in-One Image Restoration
• PreP-OCR: A Complete Pipeline for Document Image Restoration and Enhanced OCR Accuracy
• Dual Prompting for Diverse Count-level PET Denoising
• A Unified Solution to Video Fusion: From Multi-Frame Learning to Benchmarking
• Unaligned RGB Guided Hyperspectral Image Super-Resolution with Spatial-Spectral Concordance
• Freqformer: Image-Demoiréing Transformer via Efficient Frequency Decomposition
• Benchmarking Endoscopic Surgical Image Restoration and Beyond
• Efficient Degradation-agnostic Image Restoration via Channel-Wise Functional Decomposition and Manifold Regularization
• Pixel Ignores, Superpixel Sees: Adverse Weather Image Restoration via Semantic-Center SSM
• SW-ViT: A Spatio-Temporal Vision Transformer Network with Post Denoiser for Sequential Multi-Push Ultrasound Shear Wave Elastography
• RestoreVAR: Visual Autoregressive Generation for All-in-One Image Restoration
• MoCRA: Mixture of Compositional Rank-1 Atoms for 4K All-in-One Video Restoration
• SpikeRestormer: Towards Energy-Efficient All-in-One Image Restoration via Unified Event Reasoning
• Instruct2See: Learning to Remove Any Obstructions Across Distributions
• MODEM: A Morton-Order Degradation Estimation Mechanism for Adverse Weather Image Recovery
• When Extreme Darkness Meets Motion Blur: MeanFlow for Unified RAW Restoration
• Loop-Mamba: A Loop Mamba with Degradation-Aware and Shared Memory for Old Photo Restoration
• DerainSplat: Feed-Forward Clean 3D Gaussian Splatting from Sparse Rainy Views
• Deep Learning-Driven Ultra-High-Definition Image Restoration: A Survey
• Semi-Supervised State-Space Model with Dynamic Stacking Filter for Real-World Video Deraining
• Fast and Accurate Image Restoration and Generation with Rank Enhanced Linear Attention
• Projection-Based Correction for Enhancing Deep Inverse Networks
• MHANet: Multi-scale Hybrid Attention Network for Auditory Attention Detection
• Lightweight and Interpretable Transformer via Mixed Graph Algorithm Unrolling for Traffic Forecast
• Degradation-Aware Feature Perturbation for All-in-One Image Restoration
• RGB-to-Polarization Estimation: A New Task and Benchmark Study
• SMFusion: Semantic-Preserving Fusion of Multimodal Medical Images for Enhanced Clinical Diagnosis
• NTIRE 2025 Challenge on Efficient Burst HDR and Restoration: Datasets, Methods, and Results
• DiCo: Revitalizing ConvNets for Scalable and Efficient Diffusion Modeling
• Practical Noise Modeling for SPAD Intensity Imaging
• PDE: Gene Effect Inspired Parameter Dynamic Evolution for Low-light Image Enhancement
• UnfoldIR: Rethinking Deep Unfolding Network in Illumination Degradation Image Restoration
• A review of advancements in low-light image enhancement using deep learning
• Multi-Scale Target-Aware Representation Learning for Fundus Image Enhancement
• CMAWRNet: Multiple Adverse Weather Removal via a Unified Quaternion Neural Architecture
• Vision Mamba in Remote Sensing: A Comprehensive Survey of Techniques, Applications and Outlook
• Noise Modeling in One Hour: Minimizing Preparation Efforts for Self-supervised Low-Light RAW Image Denoising
• DGSolver: Diffusion Generalist Solver with Universal Posterior Sampling for Image Restoration
• A Unified Resolution-Conditioned Framework for Orthogonal Line-Scanning Image Fusion
• DPMambaIR: All-in-One Image Restoration via Degradation-Aware Prompt State Space Model
• RouteWinFormer: A Route-Window Transformer for Middle-range Attention in Image Restoration
• Cross Paradigm Representation and Alignment Transformer for Image Deraining
• DSDNet: Raw Domain Demoiréing via Dual Color-Space Synergy
• MoBGS: Motion Deblurring Dynamic 3D Gaussian Splatting for Blurry Monocular Video
• Structure-guided Diffusion Transformer for Low-Light Image Enhancement
• Frequency-domain Learning with Kernel Prior for Blind Image Deblurring
• NTIRE 2025 Challenge on Image Super-Resolution (×4): Methods and Results
• Any Image Restoration via Efficient Spatial-Frequency Degradation Adaptation
• Towards Scale-Aware Low-Light Enhancement via Structure-Guided Transformer Design
• OBIFormer: A Fast Attentive Denoising Framework for Oracle Bone Inscriptions
• Multiscale Tensor Summation Factorization as a New Neural Network Layer (MTS Layer) for Multidimensional Data Processing
• NTIRE 2025 Challenge on Day and Night Raindrop Removal for Dual-Focused Images: Methods and Results
• AdaQual-Diff: Diffusion-Based Image Restoration via Adaptive Quality Prompting
• The Tenth NTIRE 2025 Image Denoising Challenge Report
• NTIRE 2025 Challenge on Event-Based Image Deblurring: Methods and Results
• Flow-Map Distillation on Relation Manifolds for Image Restoration
• An Efficient and Mixed Heterogeneous Model for Image Restoration
• Lightweight Medical Image Restoration via Integrating Reliable Lesion-Semantic Driven Prior
• Enhancing Image Restoration through Learning Context-Rich and Detail-Accurate Features
• VibrantLeaves: A principled parametric image generator for training deep restoration models
• The Tenth NTIRE 2025 Efficient Super-Resolution Challenge Report
• Pseudo-Label Guided Real-World Image De-weathering: A Learning Framework with Imperfect Supervision
• Low-Light Image Enhancement using Event-Based Illumination Estimation
• Mitigating Long-tail Distribution in Oracle Bone Inscriptions: Dataset, Model, and Benchmark
• A Hybrid Fully Convolutional CNN-Transformer Model for Inherently Interpretable Disease Detection from Retinal Fundus Images
• X-DECODE: EXtreme Deblurring with Curriculum Optimization and Domain Equalization
• PIDSR: Complementary Polarized Image Demosaicing and Super-Resolution
• FANeRV: Frequency Separation and Augmentation based Neural Representation for Video
• Distilling Textual Priors from LLM to Efficient Image Fusion
• DA2Diff: Exploring Degradation-aware Adaptive Diffusion Priors for All-in-One Weather Restoration
• Feature Importance-Aware Deep Joint Source-Channel Coding for Computationally Efficient and Adjustable Image Transmission
• Bridging Knowledge Gap Between Image Inpainting and Large-Area Visible Watermark Removal
• DSwinIR: Rethinking Window-based Attention for Image Restoration
• Adaptive Frequency Enhancement Network for Remote Sensing Image Semantic Segmentation

Related

• Restormer: Efficient Transformer for High-Resolution Image Restoration
• Efficient Transformer for High Resolution Image Motion Deblurring
• Accurate Image Restoration with Attention Retractable Transformer
• MB-TaylorFormer V2: Improved Multi-branch Linear Transformer Expanded by Taylor Formula for Image Restoration
• Uformer: A General U-Shaped Transformer for Image Restoration