CVPR2025｜底层视觉(超分辨率，图像恢复，去雨，去雾，去模糊，去噪等)相关论文汇总（附论文链接/开源代码）【持续更新】

整理汇总下2025年底层视觉(Low-Level Vision)相关的论文和代码，括超分辨率，图像去雨，图像去雾，去模糊，去噪，图像恢复，图像增强，图像去摩尔纹，图像修复，图像质量评价，插帧，图像/视频压缩等任务，具体如下。

最新修改版本会首先更新在Github，欢迎star，fork和PR~
也欢迎对底层视觉任务感兴趣的朋友一块更新～
Github：Awesome-CVPR2025-Low-Level-Vision
知乎：https://zhuanlan.zhihu.com/p/27720923412
参考或转载请注明出处

CVPR2025官网：https://cvpr.thecvf.com/Conferences/2025

CVPR接收论文列表：https://cvpr.thecvf.com/Conferences/2025/AcceptedPapers

CVPR完整论文库：

开会时间：2025月6月11日-2025月6月15日

论文接收公布时间：2025年2月27日

【Contents】

• 1.超分辨率(Super-Resolution)
• 2.图像去雨(Image Deraining)
• 3.图像去雾(Image Dehazing)
• 4.去模糊(Deblurring)
• 5.去噪(Denoising)
• 6.图像恢复(Image Restoration)
• 7.图像增强(Image Enhancement)
• 8.图像修复(Inpainting)
• 9.高动态范围成像(HDR Imaging)
• 10.图像质量评价(Image Quality Assessment)
• 11.插帧(Frame Interpolation)
• 12.视频/图像压缩(Video/Image Compression)
• 13.压缩图像质量增强(Compressed Image Quality Enhancement)
• 14.图像去反光(Image Reflection Removal)
• 15.图像去阴影(Image Shadow Removal)
• 16.图像上色(Image Colorization)
• 17.图像和谐化(Image Harmonization)
• 18.视频稳相(Video Stabilization)
• 19.图像融合(Image Fusion)
• 20.其他任务(Others)

1.超分辨率(Super-Resolution)

Adaptive Dropout: Unleashing Dropout across Layers for Generalizable Image Super-Resolution

• Paper:
• Code: https://github.com/xuhang07/Adpative-Dropout

ADD: A General Attribution-Driven Data Augmentation Framework for Boosting Image Super-Resolution

• Paper:
• Code:

Adversarial Diffusion Compression for Real-World Image Super-Resolution

• Paper: https://arxiv.org/abs/2411.13383
• Code: https://github.com/Guaishou74851/AdcSR

Arbitrary-steps Image Super-resolution via Diffusion Inversion

• Paper: https://arxiv.org/abs/2412.09013
• Code: https://github.com/zsyOAOA/InvSR

Augmenting Perceptual Super-Resolution via Image Quality Predictors

• Paper:
• Code:

Auto-Enocded Supervision for Perceptual Image Super-Resolution

• Paper: https://arxiv.org/abs/2412.00124
• Code: https://github.com/2minkyulee/AESOP-Auto-Encoded-Supervision-for-Perceptual-Image-Super-Resolution

AutoLUT: LUT-Based Image Super-Resolution with Automatic Sampling and Adaptive Residual Learning

• Paper: https://arxiv.org/abs/2503.01565
• Code: https://github.com/SuperKenVery/AutoLUT

BF-STVSR: B-Splines and Fourier-Best Friends for High Fidelity Spatial-Temporal Video Super-Resolution

• Paper: https://arxiv.org/abs/2501.11043
• Code: https://github.com/Eunjnnn/bfstvsr

CATANet: Efficient Content-Aware Token Aggregation for Lightweight Image Super-Resolution

• Paper: https://arxiv.org/abs/2503.06896
• Code: https://github.com/EquationWalker/CATANet

Decoupling Fine Detail and Global Geometry for Compressed Depth Map Super-Resolution

• Paper: https://arxiv.org/abs/2411.03239
• Code:

DifIISR: Diffusion Model with Gradient Guidance for Infrared Image Super-Resolution

• Paper: https://www.arxiv.org/abs/2503.01187
• Code: https://github.com/zirui0625/DifIISR

DORNet: A Degradation Oriented and Regularized Network for Blind Depth Super-Resolution

• Paper: https://arxiv.org/abs/2410.11666
• Code: https://github.com/yanzq95/dornet

Edge-SD-SR: Low Latency and Parameter Efficient On-device Super-Resolution with Stable Diffusion via Bidirectional Conditioning

• Paper: https://arxiv.org/abs/2412.06978
• Code:

Efficient Video Super-Resolution for Real-time Rendering with Decoupled G-buffer Guidance

• Paper:
• Code: https://github.com/sunny2109/RDG

EvEnhancer: Empowering Effectiveness, Efficiency and Generalizability for Continuous Space-Time Video Super-Resolution with Events

• Paper:
• Code:

Event-based Video Super-Resolution via State Space Models

• Paper:
• Code:

Exploring Semantic Feature Discrimination for Perceptual Image Super-Resolution and Opinion-Unaware No-Reference Image Quality Assessment

• Paper:
• Code:

FaithDiff: Unleashing Diffusion Priors for Faithful Image Super-resolution

• Paper: https://arxiv.org/abs/2411.18824
• Code: https://github.com/JyChen9811/FaithDiff

HIIF: Hierarchical Encoding based Implicit Image Function for Continuous Super-resolution

• Paper: https://arxiv.org/abs/2412.03748v1
• Code:

Latent space Super-Resolution for Higher-Resolution Image Generation with Diffusion Models

• Paper:
• Code:

PassionSR: Post-Training Quantization with Adaptive Scale in One-Step Diffusion based Image Super-Resolution

• Paper: https://arxiv.org/abs/2411.17106
• Code: https://github.com/libozhu03/PassionSR

PatchVSR: Breaking Video Diffusion Resolution Limits with Patch-wise Video Super-Resolution

• Paper:
• Code:

PIDSR: Complementary Polarized Image Demosaicing and Super-Resolution

• Paper:
• Code:

Pixel-level and Semantic-level Adjustable Super-resolution: A Dual-LoRA Approach

• Paper: https://arxiv.org/abs/2412.03017
• Code: https://github.com/csslc/PiSA-SR

Progressive Focused Transformer for Single Image Super-Resolution

• Paper:
• Code:

QMambaBSR: Burst Image Super-Resolution with Query State Space Model

• Paper: https://arxiv.org/abs/2408.08665
• Code:

S2Gaussian: Sparse-View Super-Resolution 3D Gaussian Splatting

• Paper: https://arxiv.org/abs/2503.04314v1
• Code:

Self-supervised ControlNet with Spatio-Temporal Mamba for Real-world Video Super-resolution

• Paper:
• Code:

The Power of Context: How Multimodality Improves Image Super-Resolution

• Paper: https://arxiv.org/abs/2503.14503
• Code:

TSD-SR: One-Step Diffusion with Target Score Distillation for Real-World Image Super-Resolution

• Paper: https://arxiv.org/abs/2411.18263
• Code: https://github.com/Microtreei/TSD-SR

TSP-Mamba: The Travelling Salesman Problem Meets Mamba for Image Super-resolution and Beyond

• Paper:
• Code:

Uncertainty-guided Perturbation for Image Super-Resolution Diffusion Model

• Paper:
• Code:

VideoGigaGAN: Towards Detail-rich Video Super-Resolution

• Paper: https://arxiv.org/abs/2404.12388
• Code:

Volume Tells: Dual Cycle-Consistent Diffusion for 3D Fluorescence Microscopy De-noising and Super-Resolution

• Paper: https://arxiv.org/abs/2503.02261
• Code:

2.图像去雨(Image Deraining)

Channel Consistency Prior and Self-Reconstruction Strategy Based Unsupervised Image Deraining

• Paper:
• Code:

Semi-Supervised State-Space Model with Dynamic Stacking Filter for Real-World Video Deraining

• Paper:
• Code:

3.图像去雾(Image Dehazing)

CoA: Towards Real Image Dehazing via Compression-and-Adaptation

• Paper:
• Code: https://github.com/fyxnl/COA

Iterative Predictor-Critic Code Decoding for Real-World Image Dehazing

• Paper: https://arxiv.org/abs/2503.13147
• Code: https://github.com/Jiayi-Fu/IPC-Dehaze

Learning Hazing to Dehazing: Towards Realistic Haze Generation for Real-World Image Dehazing

• Paper:
• Code:

Tokenize Image Patches: Global Context Fusion for Effective Haze Removal in Large Images

• Paper:
• Code1: https://github.com/CastleChen339/DehazeXL
• Code2: https://github.com/fengyanzi/DehazingAttributionMap

4.去模糊(Deblurring)

A Polarization-aided Transformer for Image Deblurring via Motion Vector Decomposition

• Paper:
• Code:

Deblurring Low-light Images via Event-guided Hourglass Network

• Paper:
• Code:

DynaMoDe-NeRF: Motion-aware Deblurring Neural Radiance Field for Dynamic Scenes

• Paper:
• Code:

DiET-GS: Diffusion Prior and Event Stream-Assisted Motion Deblurring 3D Gaussian Splatting

• Paper:
• Code:

Diffusion-based Event Generation for High-Quality Image Deblurring

• Paper:
• Code:

Efficient Visual State Space Model for Image Deblurring

• Paper:
• Code:

Exploiting Deblurring Networks for Radiance Fields

• Paper:
• Code:

Gyro-based Neural Single Image Deblurring

• Paper:
• Code:

Parameterized Blur Kernel Prior Learning for Local Motion Deblurring

• Paper:
• Code:

Quad-Pixel Image Defocus Deblurring: A New Benchmark and Model

• Paper:
• Code:

5.去噪(Denoising)

All-Optical Nonlinear Diffractive Deep Network for Ultrafast Image Denoising

• Paper:
• Code:

BEVDiffuser: Plug-and-Play Diffusion Model for BEV Denoising with Ground-Truth Guidance

• Paper: https://arxiv.org/abs/2502.19694
• Code:

Blind-Spot Real-world Image Denoising via Implicit Neural Pixel Resampling

• Paper:
• Code:

Classic Video Denoising in a Machine Learning World: Robust, Fast, and Controllable

• Paper:
• Code:

Complementary Advantages: Exploiting Cross-Field Frequency Correlation for NIR-Assisted Image Denoising

• Paper:
• Code:

DnLUT: Ultra-Efficient Color Image Denoising via Channel-Aware Lookup Tables

• Paper:
• Code: https://github.com/Stephen0808/DnLUT

Noise Modeling in One Hour: Minimizing Preparation Efforts for Self-supervised Low-Light RAW Image Denoising

• Paper:
• Code:

Patient-Level Anatomy Meets Scanning-Level Physics: Personalized Federated Low-Dose CT Denoising Empowered by Large Language Model

• Paper:
• Code:

Positive2Negative: Breaking the Information-Lossy Barrier in Self-Supervised Single Image Denoising

• Paper:
• Code:

Rethinking Reconstruction and Denoising in the Dark: New Perspective, General Architecture and Beyond

• Paper:
• Code:

Rotation-Equivariant Self-Supervised Method in Image Denoising

• Paper:
• Code:

SuperPC: A Single Diffusion Model for Point Cloud Completion, Upsampling, Denoising, and Colorization

• Paper:
• Code:

6.图像恢复(Image Restoration)

ACL: Activating Capability of Linear Attention for Image Restoration

• Paper:
• Code:

Acquire and then Adapt: Squeezing out Text-to-Image Model for Image Restoration

• Paper:
• Code:

Adapting Text-to-Image Generation with Feature Difference Instruction for Generic Image Restoration

• Paper:
• Code:

A Regularization-Guided Equivariant Approach for Image Restoration

• Paper:
• Code:

Complexity Experts are Task-Discriminative Learners for Any Image Restoration

• Paper: https://arxiv.org/abs/2411.18466
• Code: https://github.com/eduardzamfir/MoCE-IR

DarkIR: Robust Low-Light Image Restoration

• Paper: https://arxiv.org/abs/2412.13443v1
• Code: https://github.com/cidautai/DarkIR

Degradation-Aware Feature Perturbation for All-in-One Image Restoration

• Paper:
• Code:

Dual Prompting for Image Restoration across Full-Scene with Diffusion Transformers

• Paper:
• Code:

Dynamic Content Prediction with Motion-aware Priors for Blind Face Video Restoration

• Paper:
• Code:

FiRe: Fixed-points of Restoration Priors for Solving Inverse Problems

• Paper: https://arxiv.org/abs/2411.18970
• Code:

From Zero to Detail: Deconstructing Ultra-High-Definition Image Restoration from Progressive Spectral Perspective

• Paper: https://arxiv.org/abs/2503.13165v1
• Code:

GenDeg: Diffusion-Based Degradation Synthesis for Generalizable All-in-One Image Restoration

• Paper: https://arxiv.org/abs/2411.17687v1
• Code: https://github.com/sudraj2002/GenDeg

Hazy Low-Quality Satellite Video Restoration Via Learning Optimal Joint Degradation Patterns and Continuous-Scale Super-Resolution Reconstruction

• Paper:
• Code:

Inverting Flow for Image Restoration

• Paper:
• Code:

JarvisIR: Elevating Autonomous Driving Perception with Intelligent Image Restoration

• Paper:
• Code:

LP-Diff: Towards Improved Restoration of Real-World Degraded License Plate

• Paper:
• Code:

MaIR: A Locality- and Continuity-Preserving Mamba for Image Restoration

• Paper: https://arxiv.org/abs/2412.20066
• Code: https://github.com/XLearning-SCU/2025-CVPR-MaIR

Making Old Film Great Again: Degradation-aware State Space Model for Old Film Restoration

• Paper:
• Code:

MambaIRv2: Attentive State Space Restoration

• Paper: https://arxiv.org/abs/2411.15269
• Code: https://github.com/csguoh/MambaIR

Navigating Image Restoration with VAR’s Distribution Alignment Prior

• Paper:
• Code:

OSDFace: One-Step Diffusion Model for Face Restoration

• Paper: https://arxiv.org/abs/2411.17163
• Code: https://github.com/jkwang28/OSDFace

Plug-and-Play Proximal Restoration Priors for Single-Pixel Imaging

• Paper:
• Code:

SeedVR: Seeding Infinity in Diffusion Transformer Towards Generic Video Restoration

• Paper: https://arxiv.org/abs/2501.01320v1
• Code:

Sparse Image Sets Restoration with Multi-View Diffusion Model

• Paper: https://arxiv.org/abs/2503.14463v1
• Code:

SVFR: A Unified Framework for Generalized Video Face Restoration

• Paper: https://arxiv.org/abs/2501.01235
• Code: https://github.com/wangzhiyaoo/SVFR

Reconciling Stochastic and Deterministic Strategies for Zero-shot Image Restoration using Diffusion Model in Dual

• Paper: https://arxiv.org/abs/2503.01288
• Code: https://github.com/ChongWang1024/RDMD

UHD-processer: Unified UHD Image Restoration with Progressive Frequency Learning and Degradation-aware Prompts

• Paper:
• Code:

UniRestore: Unified Perceptual and Task-Oriented Image Restoration Model Using Diffusion Prior

• Paper: https://arxiv.org/abs/2501.13134
• Code:

URWKV: Unified RWKV Model with Multi-state Perspective for Low-light Image Restoration

• Paper:
• Code:

Visual-Instructed Degradation Diffusion for All-in-One Image Restoration

• Paper:
• Code:

VolFormer: Explore More Comprehensive Cube Interaction for Hyperspectral Image Restoration and Beyond

• Paper:
• Code:

Zero-Shot Image Restoration Using Few-Step Guidance of Consistency Models (and Beyond)

• Paper: https://arxiv.org/abs/2412.20596
• Code: https://github.com/tirer-lab/CM4IR

7.图像增强(Image Enhancement)

3DEnhancer: Consistent Multi-View Diffusion for 3D Enhancement

• Paper: https://arxiv.org/abs/2412.18565
• Code: https://github.com/Luo-Yihang/3DEnhancer

Efficient Diffusion as Low Light Enhancer

• Paper: https://arxiv.org/abs/2410.12346
• Code: https://github.com/lgz-0713/ReDDiT

Efficient Video Face Enhancement with Enhanced Spatial-Temporal Consistency

HVI: A New Color Space for Low-light Image Enhancement

• Paper: https://arxiv.org/abs/2411.15269
• Code: https://github.com/Fediory/HVI-CIDNet

Noise Calibration and Spatial-Frequency Interactive Network for STEM Image Enhancement

• Paper:
• Code:

8.图像修复(Inpainting)

3D Gaussian Inpainting with Depth-Guided Cross-View Consistency

A${}^{\text{T}}$A: Adaptive Transformation Agent for Text-Guided Subject-Position Variable Background Inpainting

AuraFusion360: Augmented Unseen Region Alignment for Reference-based 360° Unbounded Scene Inpainting

DefectFill: Realistic Defect Generation with Inpainting Diffusion Model for Visual Inspection

HomoGen: Enhanced Video Inpainting via Homography Propagation and Diffusion

IMFine: 3D Inpainting via Geometry-guided Multi-view Refinement

Instant3dit: Multiview Inpainting for Fast Editing of 3D Objects

Large-Scale Text-to-Image Model with Inpainting is a Zero-Shot Subject-Driven Image Generator

MTADiffusion: Mask Text Alignment Diffusion Model for Object Inpainting

RAD: Region-Aware Diffusion Models for Image Inpainting

Self-Supervised Large Scale Point Cloud Completion for Archaeological Site Restoration

SVDC: Consistent Direct Time-of-Flight Video Depth Completion with Frequency Selective Fusion

• Paper: https://www.arxiv.org/abs/2503.01257
• Code: https://github.com/Lan1eve/SVDC

Towards Context-Stable and Hue-Consistent Image Inpainting

TurboFill: Adapting Few-step Text-to-image Model for Fast Image Inpainting

VideoRepainter: Creative Video Inpainting with Keyframe Reference

9.高动态范围成像(HDR Imaging)

UltraFusion: Ultra High Dynamic Imaging using Exposure Fusion

• Paper: https://arxiv.org/abs/2501.11515
• Code:

10.图像质量评价(Image Quality Assessment)

Exploring Semantic Feature Discrimination for Perceptual Image Super-Resolution and Opinion-Unaware No-Reference Image Quality Assessment

• Paper:
• Code:

FineVQ: Fine-Grained User Generated Content Video Quality Assessment

• Paper: https://arxiv.org/abs/2412.19238
• Code: https://github.com/IntMeGroup/FineVQ

Teaching Large Language Models to Regress Accurate Image Quality Scores using Score Distribution

• Paper: https://arxiv.org/abs/2501.11561
• Code: https://github.com/zhiyuanyou/DeQA-Score

Toward Generalized Image Quality Assessment: Relaxing the Perfect Reference Quality Assumption

• Paper: https://arxiv.org/abs/2503.11221
• Code: https://github.com/ChrisDud0257/AFINE

11.插帧(Frame Interpolation)

BiM-VFI: directional Motion Field-Guided Frame Interpolation for Video with Non-uniform Motions

• Paper: https://arxiv.org/abs/2412.11365
• Code:

Generative Inbetweening through Frame-wise Conditions-Driven Video Generation

• Paper: https://arxiv.org/abs/2412.11755
• Code: https://github.com/Tian-one/FCVG

12.视频/图像压缩(Video/Image Compression)

ECVC: Exploiting Non-Local Correlations in Multiple Frames for Contextual Video Compression

• Paper: https://arxiv.org/abs/2410.09706
• Code: https://github.com/JiangWeibeta/ECVC

Multirate Neural Image Compression with Adaptive Lattice Vector Quantization

• Paper:
• Code:

Test-Time Fine-Tuning of Image Compression Models for Multi-Task Adaptability

• Paper:
• Code: https://github.com/JiangWeibeta/ECVC

13.压缩图像/视频质量增强(Compressed Image/Video Quality Enhancement)

Plug-and-Play Versatile Compressed Video Enhancement

RivuletMLP: An MLP-based Architecture for Efficient Compressed Video Quality Enhancement

14.图像去反光(Image Reflection Removal)

DL2G: Degradation-guided Local-to-Global Restoration for Eyeglass Reflection Removal

Reversible Decoupling Network for Single Image Reflection Removal

• Paper: https://arxiv.org/abs/2410.08063
• Code: https://github.com/lime-j/RDNet

15.图像去阴影(Image Shadow Removal)

16.图像上色(Image Colorization)

SuperPC: A Single Diffusion Model for Point Cloud Completion, Upsampling, Denoising, and Colorization

• Paper:
• Code:

17.图像和谐化(Image Harmonization)

18.视频稳相(Video Stabilization)

19.图像融合(Image Fusion)

DCEvo: Discriminative Cross-dimensional Evolutionary Learning for Infrared and Visible Image Fusion

• Paper:
• Code: https://github.com/Beate-Suy-Zhang/DCEvo

One Model for ALL: Low-Level Task Interaction Is a Key to Task-Agnostic Image Fusion

• Paper: https://arxiv.org/abs/2502.19854
• Code: https://github.com/AWCXV/GIFNet

20.其他任务(Others)

Binarized Semantic Mamba-Transformer for Lightweight Quad Bayer HybridEVS Demosaicing

• Paper:
• Code: https://github.com/Clausy9/BMTNet

Continuous Adverse Weather Removal via Degradation-Aware Distillation

• Paper:
• Code:

Point Cloud Upsampling Using Conditional Diffusion Module with Adaptive Noise Suppression

• Paper:
• Code: https://github.com/Baty2023/PDANS

U-Know-DiffPAN: An Uncertainty-aware Knowledge Distillation Diffusion Framework with Details Enhancement for PAN-Sharpening

• Paper: https://arxiv.org/abs/2412.06243v1
• Code:

持续更新~