Awesome-Deblurring-Resources

August 30, 2024 · View on GitHub

A curated list of research papers and datasets related to image and video deblurring.

Overview

2024 Papers
2023 Papers
2022 Papers
2021 Papers
2020 Papers
2019 Papers
Datasets

2024 Papers

Venue	Paper	Link
arxiv	Blind Image Deblurring using FFT-ReLU with Deep Learning Pipeline Integration	Code
arxiv	Fast Diffusion EM: a diffusion model for blind inverse problems with application to deconvolution	FastDiffusionEM
SPIE	Estimation of motion blur kernel parameters using regression convolutional neural networks	RegressionBlur
CVPR	A Unified Framework for Microscopy Defocus Deblur with Multi-Pyramid Transformer and Contrastive Learning	MPT-CataBlur
CVPR	AdaRevD: Adaptive Patch Exiting Reversible Decoder Pushes the Limit of Image Deblurring	AdaRevD
CVPR	Blur2Blur: Blur Conversion for Unsupervised Image Deblurring on Unknown Domains	Blur2Blur
CVPR	Fourier Priors-Guided Diffusion for Zero-Shot Joint Low-Light Enhancement and Deblurring	FourierDiff
CVPR	ID-Blau: Image Deblurring by Implicit Diffusion-based reBLurring AUgmentation	ID-Blau
CVPR	LDP: Language-driven Dual-Pixel Image Defocus Deblurring Network	LDP
CVPR	Mitigating Motion Blur in Neural Radiance Fields with Events and Frames	EvDeblurNeRF
CVPR	Motion-adaptive Separable Collaborative Filters for Blind Motion Deblurring	MISCFilter
CVPR	Motion Blur Decomposition with Cross-shutter Guidance	dualBR
CVPR	Spike-guided Motion Deblurring with Unknown Modal Spatiotemporal Alignment	UaSDN
CVPR	Blur-aware Spatio-temporal Sparse Transformer for Video Deblurring	BSSTNet
CVPR	Unsupervised Blind Image Deblurring Based on Self-Enhancement	-
CVPR	Real-World Efficient Blind Motion Deblurring via Blur Pixel Discretization	-
CVPR	EVS-assisted Joint Deblurring Rolling-Shutter Correction and Video Frame Interpolation through Sensor Inverse Modeling	-
CVPR	Latency Correction for Event-guided Deblurring and Frame Interpolation	-
CVPR	Frequency-aware Event-based Video Deblurring for Real-World Motion Blur	-
arXiv	Gyroscope-Assisted Motion Deblurring Network	-
arXiv	Gyro-based Neural Single Image Deblurring	-
ECCV	BAD-Gaussians: Bundle Adjusted Deblur Gaussian Splatting	BAD-Gaussians
ECCV	BeNeRF: Neural Radiance Fields from a Single Blurry Image and Event Stream	BeNeRF
ECCV	Blind image deblurring with noise-robust kernel estimation	BD_noise_robust_kernel_estimation
ECCV	Domain-adaptive Video Deblurring via Test-time Blurring	DADeblur
ECCV	Gaussian Splatting on the Move: Blur and Rolling Shutter Compensation for Natural Camera Motion	3dgs-deblur
ECCV	Towards Real-world Event-guided Low-light Video Enhancement and Deblurring	ELEDNet
ECCV	UniINR: Event-guided Unified Rolling Shutter Correction, Deblurring, and Interpolation	UniINR

2023 Papers

Venue	Paper	Link
ICML	GibbsDDRM: A Partially Collapsed Gibbs Sampler for Solving Blind Inverse Problems with Denoising Diffusion Restoration	GibbsDDRM
IJCV	Blind Image Deblurring with Unknown Kernel Size and Substantial Noise	Blind Image Deblurring
TIP	INFWIDE: Image and Feature Space Wiener Deconvolution Network for Non-blind Image Deblurring in Low-Light Conditions	INFWIDE
AAAI	Real-World Deep Local Motion Deblurring	ReLoBlur
ICCV	Multi-scale Residual Low-Pass Filter Network for Image Deblurring	-
TCSVT	Multi-Scale Frequency Separation Network for Image Deblurring	MSFS-Net
ICML	IRNeXt: Rethinking Convolutional Network Design for Image Restoration	IRNeXt
CVPR	Structured Kernel Estimation for Photon-Limited Deconvolution	structured-kernel-cvpr23
CVPR	Blur Interpolation Transformer for Real-World Motion from Blur	BiT
CVPR	Neumann Network with Recursive Kernels for Single Image Defocus Deblurring	NRKNet
CVPR	Efficient Frequency Domain-based Transformers for High-Quality Image Deblurring	FFTformer
CVPR	Hybrid Neural Rendering for Large-Scale Scenes with Motion Blur	HybridNeuralRendering
CVPR	Self-Supervised Non-Uniform Kernel Estimation With Flow-Based Motion Prior for Blind Image Deblurring	UFPDeblur
CVPR	Uncertainty-Aware Unsupervised Image Deblurring with Deep Residual Prior	UAUDeblur
CVPR	K3DN: Disparity-Aware Kernel Estimation for Dual-Pixel Defocus Deblurring	-
CVPR	Self-Supervised Blind Motion Deblurring With Deep Expectation Maximization	-
CVPR	HyperCUT: Video Sequence from a Single Blurry Image using Unsupervised Ordering	HyperCUT
CVPR	Deep Discriminative Spatial and Temporal Network for Efficient Video Deblurring	DSTNet
AAAI	Dual-Domain Attention for Image Deblurring	DDANet
AAAI	Real-World Deep Local Motion Deblurring	ReLoBlur
AAAI	Learning Single Image Defocus Deblurring with Misaligned Training Pairs	JDRL
AAAI	Intriguing Findings of Frequency Selection for Image Deblurring	DeepRFT
AAAI	Learnable Blur Kernel for Single-Image Defocus Deblurring in the Wild	-
ICCV	Multiscale Structure Guided Diffusion for Image Deblurring	-
ICCV	Single Image Defocus Deblurring via Implicit Neural Inverse Kernels	INIKNet
ICCV	Single Image Deblurring with Row-dependent Blur Magnitude	RSS-T
ICCV	Non-Coaxial Event-Guided Motion Deblurring with Spatial Alignment	-
ICCV	Generalizing Event-Based Motion Deblurring in Real-World Scenarios	GEM
ICCV	Exploring Temporal Frequency Spectrum in Deep Video Deblurring	-
NeurIPS	Hierarchical Integration Diffusion Model for Realistic Image Deblurring	HI-Diff
NeurIPS	Enhancing Motion Deblurring in High-Speed Scenes with Spike Streams	-

2022 Papers

Venue	Paper	Link
ECCVW	MSSNet: Multi-Scale-Stage Network for Single Image Deblurring	MSSNet
CVPRW	HINet: Half Instance Normalization Network for Image Restoration	HINet
TIP	BANet: A Blur-Aware Attention Network for Dynamic Scene Deblurring	BANet
CVPR	Learning to Deblur using Light Field Generated and Real Defocus Images	DRBNet
CVPR	Pixel Screening Based Intermediate Correction for Blind Deblurring	-
CVPR	Deblurring via Stochastic Refinement	-
CVPR	XYDeblur: Divide and Conquer for Single Image Deblurring	-
CVPR	Unifying Motion Deblurring and Frame Interpolation with Events	EVDI
CVPR	E-CIR: Event-Enhanced Continuous Intensity Recovery	E-CIR
CVPR	Multi-Scale Memory-Based Video Deblurring	MemDeblur
ECCV	Learning Degradation Representations for Image Deblurring	Learning_degradation
ECCV	Stripformer: Strip Transformer for Fast Image Deblurring	Stripformer-ECCV-2022
ECCV	Animation from Blur: Multi-modal Blur Decomposition with Motion Guidance	Animation-from-Blur
ECCV	United Defocus Blur Detection and Deblurring via Adversarial Promoting Learning	APL
ECCV	Realistic Blur Synthesis for Learning Image Deblurring	RSBlur
ECCV	Event-based Fusion for Motion Deblurring with Cross-modal Attention	EFNet
ECCV	Event-Guided Deblurring of Unknown Exposure Time Videos	UEVD_public
ECCV	Spatio-Temporal Deformable Attention Network for Video Deblurring	STDAN
ECCV	Efficient Video Deblurring Guided by Motion Magnitude	MMP-RNN
ECCV	ERDN: Equivalent Receptive Field Deformable Network for Video Deblurring	ERDN
ECCV	DeMFI: Deep Joint Deblurring and Multi-Frame Interpolation with Flow-Guided Attentive Correlation and Recursive Boosting	DeMFI
ECCV	Towards Real-World Video Deblurring by Exploring Blur Formation Process	RAWBlur

2021 Papers

Venue	Paper	Link
CVPR	Explore Image Deblurring via Encoded Blur Kernel Space	Blur-Kernel-Space-Exploring
ICCV	Rethinking Coarse-to-Fine Approach in Single Image Deblurring	MIMO-UNet
CVPR	Multi-Stage Progressive Image Restoration	MPRNet
CVPR	DeFMO: Deblurring and Shape Recovery of Fast Moving Objects	DeFMO
CVPR	ARVo: Learning All-Range Volumetric Correspondence for Video Deblurring	-
CVPR	Towards Rolling Shutter Correction and Deblurring in Dynamic Scenes	RSCD
CVPR	Digital Gimbal: End-to-end Deep Image Stabilization with Learnable Exposure Times	Digital Gimbal
ICCV	Bringing Events into Video Deblurring with Non consecutively Blurry Frames	D2Net
ICCV	Rethinking Coarse-to-Fine Approach in Single Image Deblurring	MIMO-UNet
ICCV	Single Image Defocus Deblurring Using Kernel-Sharing Parallel Atrous Convolutions	KPAC
NeurIPS	Gaussian Kernel Mixture Network for Single Image Defocus Deblurring	GKMNet

2020 Papers

Venue	Paper	Link
NeurIPS	Deep Wiener Deconvolution: Wiener Meets Deep Learning for Image Deblurring	DWDN
IEEE	Raw Image Deblurring	Raw Image Deblurring
TCSVT	A Simple Local Minimal Intensity Prior and An Improved Algorithm for Blind Image Deblurring	Deblur-PMP
ECCV	End-to-end Interpretable Learning of Non-blind Image Deblurring	CPCR
IJCV	Spatially-Adaptive Filter Units for Compact and Efficient Deep Neural Networks	DAU-ConvNet
TNNLS	Learning Deep Gradient Descent Optimization for Image Deconvolution	Learn-Optimizer-RGDN
TCSVT	Deep Convolutional-Neural-Network-Based Channel Attention for Single Image Dynamic Scene Blind Deblurring	-
CVPR	Cascaded Deep Video Deblurring Using Temporal Sharpness Prior	CDVD-TSP
CVPR	Learning Event-Based Motion Deblurring	-
CVPR	Variational-EM-Based Deep Learning for Noise-Blind Image Deblurring	VEM-NBD
CVPR	Efficient Dynamic Scene Deblurring Using Spatially Variant Deconvolution Network With Optical Flow Guided Training	-
CVPR	Deblurring by Realistic Blurring	Deblurring-by-Realistic-Blurring
CVPR	Spatially-Attentive Patch-Hierarchical Network for Adaptive Motion Deblurring	-
CVPR	Deblurring Using Analysis-Synthesis Networks Pair	-
ECCV	Efficient Spatio-Temporal Recurrent Neural Network for Video Deblurring	ESTRNN
ECCV	Multi-Temporal Recurrent Neural Networks For Progressive Non-Uniform Single Image Deblurring With Incremental Temporal Training	MTRNN
ECCV	Learning Event-Driven Video Deblurring and Interpolation	LEDVDI
ECCV	Defocus Deblurring Using Dual-Pixel Data	defocus-deblurring-dual-pixel
ECCV	Real-World Blur Dataset for Learning and Benchmarking Deblurring Algorithms	RealBlur
ECCV	OID: Outlier Identifying and Discarding in Blind Image Deblurring	OID
ECCV	Enhanced Sparse Model for Blind Deblurring	Enhanced Sparse Model

2019 Papers

Venue	Paper	Link
ICCV	DeblurGAN-v2: Deblurring (Orders-of-Magnitude) Faster and Better	DeblurGANv2
BMVC	Blind Image Deconvolution using Pretrained Generative Priors	Blind Image Deconvolution
arxiv	Efficient Blind Deblurring under High Noise Levels	High-Noise-Deblurring
CVPR	Deep Stacked Hierarchical Multi-Patch Network for Image Deblurring	DMPHN
CVPR	Dynamic Scene Deblurring with Parameter Selective Sharing and Nested Skip Connections	Deblur

Datasets

Name	Description	Link
GoPro	The GoPro dataset consists of 3,214 pairs of motion-blurred and sharp images, each with a resolution of 1,280×720 pixels, divided into 2,103 training pairs and 1,111 test pairs.	GoPro
REDS	The REalistic and Dynamic Scenes (REDS) dataset is generated from 120 fps videos, with blurry frames synthesized by merging consecutive frames, capturing realistic motion blur in dynamic scenes.	REDS
DPDD	The Dual-Pixel Defocus Deblurring (DPDD) dataset contains 500 carefully captured scenes, comprising 2000 images in total: 500 defocus-blurred images with their 1000 dual-pixel (DP) sub-aperture views and 500 corresponding all-in-focus images, all at full-frame resolution of 6720x4480 pixels.	DPDD
HIDE	The HIDE (Human-aware Image Deblurring) dataset consists of 8,422 blurred images paired with their corresponding sharp images, focusing on motion deblurring with an emphasis on human subjects, making it ideal for human-centric deblurring tasks.	HIDE
RealBlur	The RealBlur dataset consists of 4,738 pairs of images from 232 different scenes, captured in both camera raw and JPEG formats. It is divided into two subsets: RealBlur-R with raw images and RealBlur-J with JPEG images, with 3,758 training pairs and 980 test pairs in each subset.	RealBlur
CelebA	The CelebFaces Attributes dataset (CelebA) is a large-scale face attributes dataset comprising 202,599 images of 10,177 celebrities. Each image is 178×218 pixels and annotated with 40 binary labels for facial attributes like hair color, gender, and age.	CelebA
Deblur-NeRF	The Deblur-NeRF dataset focuses on two types of blur: camera motion blur and defocus blur. It includes 5 synthesized scenes for each blur type, created using Blender with multi-view cameras to simulate real data capture. For motion blur, images are rendered from interpolated camera poses, while defocus blur images are generated with depth-of-field effects. Additionally, the dataset features 20 real-world scenes—10 for each blur type—captured with a Canon EOS RP, including both manually blurred images and sharp reference images.	Deblur-NeRF
RSBlur	The RSBlur dataset offers pairs of real and synthetic blurred images, each with corresponding ground truth sharp images. It is designed to evaluate deblurring and blur synthesis methods on real-world blurred images, with training, validation, and test sets comprising 8,878, 1,120, and 3,360 blurred images, respectively.	RSBlur
ReloBlur	The ReloBlur dataset for local motion deblurring consists of 2405 blurred images with the size of 2152×1436 that are divided into 2010 training images and 395 test images. The dataset consists of pairs of a realistic locally blurred image and the corresponding ground truth sharp image that are obtained by a synchronized beam-splitting photographing system. For efficient training and testing, we also provide the resized version of ReLoBlur Dataset with the size of 538x359.	ReloBrur