整理汇总下2026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-CVPR2026-Low-Level-Vision
知乎：https://zhuanlan.zhihu.com/p/2011913167840764194
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)

xxx

• Paper:
• Code:

AlignVAR: Towards Globally Consistent Visual Autoregression for Image Super-Resolution

• Paper: https://arxiv.org/abs/2603.00589
• Code:

Bridging Fidelity-Reality with Controllable One-Step Diffusion for Image Super-Resolution

• Paper: https://arxiv.org/abs/2512.14061
• Code: https://github.com/Chanson94/CODSR

CASR: A Robust Cyclic Framework for Arbitrary Large-Scale Super-Resolution with Distribution Alignment and Self-Similarity Awareness

• Paper: https://arxiv.org/abs/2602.22159
• Code:

Compressed-Domain-Aware Online Video Super-Resolution

• Paper: https://arxiv.org/abs/2603.07694
• Code: https://github.com/yuhangwang-ai/CDA-VSR

Disentangled Textual Priors for Diffusion-based Image Super-Resolution

• Paper: https://arxiv.org/abs/2603.07430
• Code: https://github.com/JL6666JL/DTPSR

DNF-SR: Dual-Input and Negative-Aware Feature Fine-Tuning for Real-World Image Super-Resolution

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Han_DNF-SR_Dual-Input_and_Negative-Aware_Feature_Fine-Tuning_for_Real-World_Image_Super-Resolution_CVPR_2026_paper.html
• Code: https://github.com/SHH-Han/DNF-SR

DreamSR: Towards Ultra-High-Resolution Image Super-Resolution via a Receptive-Field Enhanced Diffusion Transformer

• Paper: https://arxiv.org/abs/2605.15682
• Code: https://github.com/jerrydong0219/DreamSR

DTG-Restore: Training-Free Diffusion Refinement for Generative Video Super-Resolution

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Yesiltepe_DTG-Restore_Training-Free_Diffusion_Refinement_for_Generative_Video_Super-Resolution_CVPR_2026_paper.html
• Code:

Dual Graph Regularized Deep Unfolding Network for Guided Depth Map Super-resolution

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Zhong_Dual_Graph_Regularized_Deep_Unfolding_Network_for_Guided_Depth_Map_CVPR_2026_paper.html
• Code: https://github.com/zhwzhong/LapNet

DUO-VSR: Dual-Stream Distillation for One-Step Video Super-Resolution

• Paper: https://arxiv.org/abs/2603.22271
• Code: https://github.com/cszy98/DUO-VSR

DVAR: Dynamic Visual Autoregressive Modeling for Image Super-Resolution

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Zheng_DVAR_Dynamic_Visual_Autoregressive_Modeling_for_Image_Super-Resolution_CVPR_2026_paper.html
• Code:

Edge-aware Multimodal Residual Diffusion Model for Hyperspectral Image Super-resolution

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Zhang_EMR-Diff_Edge-aware_Multimodal_Residual_Diffusion_Model_for_Hyperspectral_Image_Super-resolution_CVPR_2026_paper.html
• Code: https://github.com/luocz55/EMR-Diff

Edge-Focused Super-Resolution for Omnidirectional Images with Spherical Geometric Augmentation

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Wang_Edge-Focused_Super-Resolution_for_Omnidirectional_Images_with_Spherical_Geometric_Augmentation_CVPR_2026_paper.html
• Code:

Enhancing Unregistered Hyperspectral Image Super-Resolution via Unmixing-based Abundance Fusion Learning

• Paper: https://arxiv.org/abs/2603.07918
• Code:

FiDeSR: High-Fidelity and Detail-Preserving One-Step Diffusion Super-Resolution

• Paper: https://arxiv.org/abs/2603.02692
• Code: https://github.com/Ar0Kim/FiDeSR

FinPercep-RM: A Fine-grained Reward Model and Co-evolutionary Curriculum for RL-based Real-world Super-Resolution

• Paper: https://arxiv.org/abs/2512.22647
• Code: https://github.com/lyd-2022/FinPercep-RM

FRAMER: Frequency-Aligned Self-Distillation with Adaptive Modulation Leveraging Diffusion Priors for Real-World Image Super-Resolution

• Paper: https://arxiv.org/abs/2512.01390
• Code: https://github.com/CMLab-Korea/CVPR26-FRAMER

GDPO-SR: Group Direct Preference Optimization for One-Step Generative Image Super-Resolution

• Paper: https://arxiv.org/abs/2603.16769
• Code: https://github.com/Joyies/GDPO

Gradient Knows Best: Mixed-Precision Quantization via Gradient-Guided Bit Allocation for Super-Resolution

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Kim_Gradient_Knows_Best_Mixed-Precision_Quantization_via_Gradient-Guided_Bit_Allocation_for_CVPR_2026_paper.html
• Code:

HDW-SR: High-Frequency Guided Diffusion Model based on Wavelet Decomposition for Image Super-Resolution

• Paper: https://arxiv.org/abs/2511.13175
• Code:

IAFMNet: Information-Aware Feature Modulation for Efficient Super-Resolution

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Xu_IAFMNet_Information-Aware_Feature_Modulation_for_Efficient_Super-Resolution_CVPR_2026_paper.html
• Code:

IFCSR: Inference-Free Fidelity-Realism Control for One-Step Diffusion-based Real-World Image Super-Resolution

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Back_IFCSR_Inference-Free_Fidelity-Realism_Control_for_One-Step_Diffusion-based_Real-World_Image_Super-Resolution_CVPR_2026_paper.html
• Code:

Linear Recurrent Unit with Semantic Modulation for Image Super-Resolution

• Paper: https://openaccess.thecvf.com/content/CVPR2026F/html/Choi_Linear_Recurrent_Unit_with_Semantic_Modulation_for_Image_Super-Resolution_CVPRF_2026_paper.html
• Code: https://github.com/MingyuChoi-run/LSM

LSGQuant: Layer-Sensitivity Guided Quantization for One-Step Diffusion Real-World Video Super-Resolution

• Paper: https://arxiv.org/abs/2602.03182
• Code: https://github.com/zhengchen1999/LSGQuant

One-Step Diffusion Transformer for Controllable Real-World Image Super-Resolution

• Paper: https://arxiv.org/abs/2511.17138
• Code: https://github.com/RedMediaTech/ODTSR

Physics-Consistent Diffusion for Efficient Fluid Super-Resolution via Multiscale Residual Correction

• Paper: https://arxiv.org/abs/2603.00149
• Code: https://github.com/lizhihao2022/ReMD

Plug-and-Play Dynamic In-context Learning with Stochastic Regularization for Screen Content Image Super-Resolution

• Paper: https://openaccess.thecvf.com/content/CVPR2026F/html/Wang_Plug-and-Play_Dynamic_In-context_Learning_with_Stochastic_Regularization_for_Screen_Content_CVPRF_2026_paper.html
• Code:

PS-SR: Pseudo-Single-Step Video Super-Resolution via Speculative Diffusion

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Wu_PS-SR_Pseudo-Single-Step_Video_Super-Resolution_via_Speculative_Diffusion_CVPR_2026_paper.html
• Code:

QDM: Quadtree-Based Region-Adaptive Sparse Diffusion Models for Efficient Image Super-Resolution

• Paper: https://openaccess.thecvf.com/content/CVPR2026F/html/Yang_QDM_Quadtree-Based_Region-Adaptive_Sparse_Diffusion_Models_for_Efficient_Image_Super-Resolution_CVPRF_2026_paper.html
• Code: https://github.com/linYDTHU/QDM

RAW-Domain Degradation Models for Realistic Smartphone Super-Resolution

• Paper: https://arxiv.org/abs/2603.12493
• Code:

Remote Sensing Image Super-Resolution for Imbalanced Textures: A Texture-Aware Diffusion Framework

• Paper: https://arxiv.org/abs/2604.13994
• Code:

Restore Text First, Enhance Image Later: Two-Stage Scene Text Image Super-Resolution with Glyph Structure Guidance

• Paper: https://arxiv.org/abs/2510.21590
• Code:

Rethinking Diffusion Model-Based Video Super-Resolution: Leveraging Dense Guidance from Aligned Features

• Paper: https://arxiv.org/abs/2511.16928
• Code: https://github.com/tszssong/DGAF-VSR

SAT: Selective Aggregation Transformer for Image Super-Resolution

• Paper: https://arxiv.org/abs/2604.07994
• Code:

Spectral Super-Resolution via Adversarial Unfolding and Data-Driven Spectrum Regularization

• Paper: https://arxiv.org/abs/2603.00920
• Code:

SR3R: Rethinking Super-Resolution 3D Reconstruction With Feed-Forward Gaussian Splatting

• Paper: https://arxiv.org/abs/2602.24020
• Code:

STCDiT: Spatio-Temporally Consistent Diffusion Transformer for High-Quality Video Super-Resolution

• Paper: https://arxiv.org/abs/2511.18786
• Code: https://github.com/JyChen9811/STCDiT

TextOVSR: Text-Guided Real-World Opera Video Super-Resolution

• Paper: https://arxiv.org/abs/2603.15153
• Code: https://github.com/ChangHua0/TextOVSR

Thermal Diffusion Matters: Infrared Spatial-Temporal Video Super-Resolution through Heat Conduction Priors

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Zhou_Thermal_Diffusion_Matters_Infrared_Spatial-Temporal_Video_Super-Resolution_through_Heat_Conduction_CVPR_2026_paper.html
• Code:

Time-Aware One Step Diffusion Network for Real-World Image Super-Resolution

• Paper: https://arxiv.org/abs/2508.16557
• Code: https://github.com/zty557/TADSR

Time Without Time: Pseudo-Temporal Representation for Space-Time Super-Resolution

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Choi_Time_Without_Time_Pseudo-Temporal_Representation_for_Space-Time_Super-Resolution_CVPR_2026_paper.html
• Code:

TinySR: Shallow Diffusion Transformers for Real-World Image Super-Resolution

• Paper: https://openaccess.thecvf.com/content/CVPR2026F/html/Dong_TinySR_Shallow_Diffusion_Transformers_for_Real-World_Image_Super-Resolution_CVPRF_2026_paper.html
• Code: https://github.com/Microtreei/TinySR

Towards Real-Time Diffusion-Based Streaming Video Super-Resolution

• Paper: https://arxiv.org/abs/2510.12747
• Code: https://github.com/OpenImagingLab/FlashVSR

Toward Real-world Infrared Image Super-Resolution: A Unified Autoregressive Framework and Benchmark Dataset

• Paper: https://arxiv.org/abs/2603.04745
• Code: https://github.com/JZD151/Real-IISR

TPTransformer: Tensor-Tensor Product Transformer for Hyperspectral Image Super-Resolution

• Paper: https://openaccess.thecvf.com/content/CVPR2026F/html/Xu_TPTransformer_Tensor-Tensor_Product_Transformer_for_Hyperspectral_Image_Super-Resolution_CVPRF_2026_paper.html
• Code:

TUDSR: Twice Upsampling-Diffusion for Higher Super-Resolution

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Wu_TUDSR_Twice_Upsampling-Diffusion_for_Higher_Super-Resolution_CVPR_2026_paper.html
• Code:

UCAN: Unified Convolutional Attention Network for Expansive Receptive Fields in Lightweight Super-Resolution

• Paper: https://arxiv.org/abs/2603.11680
• Code: https://github.com/hokiyoshi/UCAN

VSRELL: A Simple Baseline for Video Super-Resolution and Enhancement in Low-Light environment

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Hui_VSRELL_A_Simple_Baseline_for_Video_Super-Resolution_and_Enhancement_in_CVPR_2026_paper.html
• Code:

VoDaSuRe: A Large-Scale Dataset Revealing Domain Shift in Volumetric Super-Resolution

• Paper: https://arxiv.org/abs/2603.23153
• Code: https://augusthoeg.github.io/VoDaSuRe/

VOSR: A Vision-Only Generative Model for Image Super-Resolution

• Paper: https://arxiv.org/abs/2604.03225
• Code: https://github.com/cswry/VOSR

2.图像去雨(Image Deraining)

UniRain: Unified Image Deraining with RAG-based Dataset Distillation and Multi-objective Reweighted Optimization

• Paper: https://arxiv.org/abs/2603.03967
• Code: https://github.com/QianfengY/UniRain

Unpaired Image Deraining Using Reward-Guided Self-Reinforcement Strategy

• Paper: https://arxiv.org/abs/2605.00719
• Code:

3.图像去雾(Image Dehazing)

Bilevel Layer-Positioning LoRA for Real Image Dehazing

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Zhang_Bilevel_Layer-Positioning_LoRA_for_Real_Image_Dehazing_CVPR_2026_paper.html
• Code: https://github.com/YanZhang-zy/BiLaLoRA

Disentanglement-wise Image Dehazing through Cross-Domain Manifold Consensus

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Lyu_Disentanglement-wise_Image_Dehazing_through_Cross-Domain_Manifold_Consensus_CVPR_2026_paper.html
• Code:

From Events to Clarity: The Event-Guided Diffusion Framework for Dehazing

• Paper: https://arxiv.org/abs/2511.11944
• Code:

HazeMatching: Dehazing Light Microscopy Images with Guided Conditional Flow Matching

• Paper: https://arxiv.org/abs/2506.22397
• Code: https://github.com/juglab/HazeMatching

Inf-Dehaze: Beyond GPU Memory Constraints for Ultra-High-Resolution Image Dehazing

• Paper: https://openaccess.thecvf.com/content/CVPR2026F/html/Yan_Inf-Dehaze_Beyond_GPU_Memory_Constraints_for_Ultra-High-Resolution_Image_Dehazing_CVPRF_2026_paper.html
• Code: https://github.com/fengyanzi/Inf-Dehaze

4.去模糊(Deblurring)

BluRef: Unsupervised Image Deblurring with Dense-Matching References

• Paper: https://arxiv.org/abs/2603.14176
• Code:

Event-based Motion Deblurring with Unpaired Data

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Cho_Event-based_Motion_Deblurring_with_Unpaired_Data_CVPR_2026_paper.html
• Code:

Gyro-based Deep Video Deblurring

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Rim_Gyro-based_Deep_Video_Deblurring_CVPR_2026_paper.html
• Code:

Motion-Aware Animatable Gaussian Avatars Deblurring

• Paper: https://arxiv.org/abs/2411.16758
• Code: https://github.com/MyNiuuu/MAD-Avatar

MSCD-GS: Motion-Separated Cooperative Deblurring Dynamic Reconstruction via Gaussian Splatting

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Liao_MSCD-GS_Motion-Separated_Cooperative_Deblurring_Dynamic_Reconstruction_via_Gaussian_Splatting_CVPR_2026_paper.html
• Code:

MVSSM: Motion-aware Visual State Space Model for Efficient Video Deblurring

• Paper: https://openaccess.thecvf.com/content/CVPR2026F/html/Zhou_MVSSM_Motion-aware_Visual_State_Space_Model_for_Efficient_Video_Deblurring_CVPRF_2026_paper.html
• Code: https://github.com/Frank-Zhou-01/MVSSM

OMoBlur: An Object Motion Blur Dataset and Benchmark for Real-World Local Motion Deblurring

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Yu_OMoBlur_An_Object_Motion_Blur_Dataset_and_Benchmark_for_Real-World_CVPR_2026_paper.html
• Code: https://github.com/yudingchuan/OMDNet

SelfHVD: Self-Supervised Handheld Video Deblurring

• Paper: https://arxiv.org/abs/2508.08605
• Code: https://github.com/cshonglei/SelfHVD

Spatio-Temporal Difference Guided Motion Deblurring with the Complementary Vision Sensor

• Paper: https://arxiv.org/abs/2604.10554
• Code:

Unblur-SLAM: Dense Neural SLAM for Blurry Inputs

• Paper: https://arxiv.org/abs/2603.26810
• Code: https://github.com/SlamMate/Unblur-SLAM

5.去噪(Denoising)

2-Shots in the Dark: Low-Light Denoising with Minimal Data Acquisition

• Paper: https://arxiv.org/abs/2512.03245
• Code:

Back to Basics: Let Denoising Generative Models Denoise

• Paper: https://arxiv.org/abs/2511.13720
• Code:

Convexity-Aware Noise Calibration: A Self-Supervised Framework for Noise-Level-Unknown Image Denoising

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Wang_Convexity-Aware_Noise_Calibration_A_Self-Supervised_Framework_for_Noise-Level-Unknown_Image_Denoising_CVPR_2026_paper.html
• Code:

Diffusion-Based sRGB Real Noise Generation via Prompt-Driven Noise Representation Learning

• Paper: https://arxiv.org/abs/2603.04870
• Code:

Efficient Real-Time Raw-to-Raw Denoising for Extreme Low-Light Ultra HD Video on Mobile Devices

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Pochimireddy_Efficient_Real-Time_Raw-to-Raw_Denoising_for_Extreme_Low-Light_Ultra_HD_Video_CVPR_2026_paper.html
• Code:

Learning to Translate Noise for Robust Image Denoising

• Paper: https://arxiv.org/abs/2412.04727
• Code: https://github.com/dhryougit/learning-to-translate-noise

LF-BVN: Blind-View Network for Self-Supervised Light Field Denoising

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Guo_LF-BVN_Blind-View_Network_for_Self-Supervised_Light_Field_Denoising_CVPR_2026_paper.html
• Code:

Next-Scale Prediction: A Self-Supervised Approach for Real-World Image Denoising

• Paper: https://arxiv.org/abs/2512.21038
• Code:

RelativeFlow: Taming Medical Image Denoising Learning with Noisy Reference

• Paper: https://arxiv.org/abs/2604.15459
• Code:

Routing on Demand: DSNet for Efficient Progressive Point Cloud Denoising

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Cheng_Routing_on_Demand_DSNet_for_Efficient_Progressive_Point_Cloud_Denoising_CVPR_2026_paper.html
• Code:

Statistical Characteristic-Guided Denoising for Rapid High-Resolution Transmission Electron Microscopy Imaging

• Paper: https://arxiv.org/abs/2603.18834
• Code: https://github.com/HeasonLee/SCGN

TM-BSN: Triangular-Masked Blind-Spot Network for Real-World Self-Supervised Denoising

• Paper: https://arxiv.org/abs/2604.04484
• Code: https://github.com/parkjun210/TM-BSN

Zero-Shot Image Denoising via Hybrid Prior-Guided Pseudo Sample Generation

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Zhao_Zero-Shot_Image_Denoising_via_Hybrid_Prior-Guided_Pseudo_Sample_Generation_CVPR_2026_paper.html
• Code:

6.图像恢复(Image Restoration)

Benchmarking Endoscopic Surgical Image Restoration and Beyond

• Paper: https://arxiv.org/abs/2505.19161
• Code:

Beyond Ground-Truth: Leveraging Image Quality Priors for Real-World Image Restoration

• Paper: https://arxiv.org/abs/2603.29773
• Code:

Beyond the Ground Truth: Enhanced Supervision for Image Restoration

• Paper: https://arxiv.org/abs/2512.03932
• Code: https://github.com/dhryougit/Beyond-the-Ground-Truth

Beyond the Static-World: Lifelong Learning for All-in-One Medical Image Restoration

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Shan_Beyond_the_Static-World_Lifelong_Learning_for_All-in-One_Medical_Image_Restoration_CVPR_2026_paper.html
• Code:

Blockwise Divide-and-Aggregate for Image Restoration using Diffusion Priors

• Paper: https://openaccess.thecvf.com/content/CVPR2026F/html/Purohit_Blockwise_Divide-and-Aggregate_for_Image_Restoration_using_Diffusion_Priors_CVPRF_2026_paper.html
• Code:

CARD: Correlation Aware Restoration with Diffusion

• Paper: https://arxiv.org/abs/2512.05268
• Code:

DEBIR: Dynamic Exposure Burst Image Restoration

• Paper: https://arxiv.org/abs/2603.21784
• Code: https://github.com/woo525/DEBIR

Degradation-Consistent Test-Time Adaptation for All-in-One Image Restoration

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Tang_Degradation-Consistent_Test-Time_Adaptation_for_All-in-One_Image_Restoration_CVPR_2026_paper.html
• Code:

DRIFT: Deep Restoration, ISP Fusion, and Tone-mapping

• Paper: https://arxiv.org/abs/2604.03402
• Code:

EpiAgent: An Agent-Centric System for Ancient Inscription Restoration

• Paper: https://arxiv.org/abs/2604.09367
• Code: https://github.com/blackprotoss/EpiAgent

Evolutionary Multi-Agent Collaboration for Real-World Video Face Restoration

• Paper: https://openaccess.thecvf.com/content/CVPR2026F/html/Tang_Evolutionary_Multi-Agent_Collaboration_for_Real-World_Video_Face_Restoration_CVPRF_2026_paper.html
• Code:

Face2Scene: Using Facial Degradation as an Oracle for Diffusion-Based Scene Restoration

• Paper: https://arxiv.org/abs/2603.16570
• Code:

FAPE-IR: Frequency-Aware Planning and Execution Framework for All-in-One Image Restoration

• Paper: https://arxiv.org/abs/2511.14099
• Code: https://github.com/Programmergg/FAPE-IR

FlowSteer: Conditioning Flow Field for Consistent Image Restoration

• Paper: https://openaccess.thecvf.com/content/CVPR2026F/html/Wickremasinghe_FlowSteer_Conditioning_Flow_Field_for_Consistent_Image_Restoration_CVPRF_2026_paper.html
• Code:

FoundIR-v2: Optimizing Pre-Training Data Mixtures for Image Restoration Foundation Model

• Paper: https://arxiv.org/abs/2512.09282
• Code: https://github.com/cschenxiang/FoundIR-v2

Gaussian Splatting-based Low-Rank Tensor Representation for Multi-Dimensional Image Recovery

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Zeng_Gaussian_Splatting-based_Low-Rank_Tensor_Representation_for_Multi-Dimensional_Image_Recovery_CVPR_2026_paper.html
• Code:

gQIR: Generative Quanta Image Reconstruction

• Paper: https://arxiv.org/abs/2602.20417
• Code: https://github.com/Aryan-Garg/gQIR

GSNR: Graph Smooth Null-Space Representation for Inverse Problems

• Paper: https://arxiv.org/abs/2602.20328
• Code:

HalluGen: Synthesizing Realistic and Controllable Hallucinations for Evaluating Image Restoration

• Paper: https://arxiv.org/abs/2512.03345
• Code:

How far have we gone in Generative Image Restoration? A study on its capability, limitations and evaluation practices

• Paper: https://openaccess.thecvf.com/content/CVPR2026F/html/Yin_How_far_have_we_gone_in_Generative_Image_Restoration_A_CVPRF_2026_paper.html
• Code:

Hybrid Agents for Image Restoration

• Paper: https://arxiv.org/abs/2503.10120
• Code:

LWTformer: A Detail-Aware, Learnable Wavelet-Transformer for Ancient Chinese Character Image Restoration

• Paper: https://openaccess.thecvf.com/content/CVPR2026F/html/Ruan_LWTformer_A_Detail-Aware_Learnable_Wavelet-Transformer_for_Ancient_Chinese_Character_Image_CVPRF_2026_paper.html
• Code: https://github.com/INWLY/LWTformer

MMDIR: Multimodal Instruction-Driven Framework for Mixed-Degradation Document Image Restoration

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Li_MMDIR_Multimodal_Instruction-Driven_Framework_for_Mixed-Degradation_Document_Image_Restoration_CVPR_2026_paper.html
• Code:

NanoSD: Edge Efficient Foundation Model for Real Time Image Restoration

• Paper: https://arxiv.org/abs/2601.09823
• Code:

Optical Tolerance-Compensated Diffusion Model for Image Restoration

• Paper: https://openaccess.thecvf.com/content/CVPR2026F/html/Dong_Optical_Tolerance-Compensated_Diffusion_Model_for_Image_Restoration_CVPRF_2026_paper.html
• Code:

PGDR-BambooSlips: Physics-Guided Multistep Deformation Reversal for Ancient Bamboo Slip Restoration

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Tang_Physics-Guided_Multistep_Deformation_Reversal_for_Ancient_Bamboo_Slip_Restoration_CVPR_2026_paper.html
• Code: https://github.com/VillanelleQQ/PGDR-BambooSlips

Q-MambaIR: Accurate Quantized Mamba for Efficient Image Restoration

• Paper: https://openaccess.thecvf.com/content/CVPR2026F/html/Chen_Q-MambaIR_Accurate_Quantized_Mamba_for_Efficient_Image_Restoration_CVPRF_2026_paper.html
• Code: https://github.com/Areache/Q-MambaIR

RDBM: Residual Diffusion Bridge Model for Image Restoration

• Paper: https://arxiv.org/abs/2510.23116
• Code: https://github.com/MiliLab/RDBM

Residual Diffusion Bridge Model for Image Restoration

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Wang_Residual_Diffusion_Bridge_Model_for_Image_Restoration_CVPR_2026_paper.html
• Code: https://github.com/MiliLab/RDBM

Restore-R1: Efficient Image Restoration Agents via Reinforcement Learning with Multimodal LLM Perceptual Feedback

• Paper: https://openaccess.thecvf.com/content/CVPR2026F/html/Lu_Restore-R1_Efficient_Image_Restoration_Agents_via_Reinforcement_Learning_with_Multimodal_CVPRF_2026_paper.html
• Code:

Restore, Assess, Repeat: A Unified Framework for Iterative Image Restoration

• Paper: https://arxiv.org/abs/2603.26385
• Code: https://github.com/SamsungLabs/RAR

Retrieve-to-Restore: Efficient All-in-One Image Restoration with a Retrieval-Based Degradation Bank

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Wang_Retrieve-to-Restore_Efficient_All-in-One_Image_Restoration_with_a_Retrieval-Based_Degradation_Bank_CVPR_2026_paper.html
• Code:

Scan Clusters, Not Pixels: A Cluster-Centric Paradigm for Efficient Ultra-high-definition Image Restoration

• Paper: https://arxiv.org/abs/2602.21917
• Code: https://github.com/5chen/C2SSM

Self-supervised Dynamic Heterogeneous Degradation Modeling for Unified Zero-Shot Image Restoration

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Hu_Self-supervised_Dynamic_Heterogeneous_Degradation_Modeling_for_Unified_Zero-Shot_Image_Restoration_CVPR_2026_paper.html
• Code:

ShiftLUT: Spatial Shift Enhanced Look-Up Tables for Efficient Image Restoration

• Paper: https://arxiv.org/abs/2603.00906
• Code: https://github.com/Sailor-t/ShiftLUT

Surgical Image Restoration Benchmark

• Paper: https://arxiv.org/abs/2505.19161
• Code: https://github.com/PJLallen/Surgical-Image-Restoration

UARE: A Unified Vision-Language Model for Image Quality Assessment, Restoration, and Enhancement

• Paper: https://arxiv.org/abs/2512.06750
• Code:

UCMNet: Uncertainty-Aware Context Memory Network for Under-Display Camera Image Restoration

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Kim_UCMNet_Uncertainty-Aware_Context_Memory_Network_for_Under-Display_Camera_Image_Restoration_CVPR_2026_paper.html
• Code: https://github.com/kdhRick2222/UCMNet

UnfoldIR: Rethinking Deep Unfolding Network in Illumination Degradation Image Restoration

• Paper: https://openaccess.thecvf.com/content/CVPR2026F/html/He_UnfoldIR_Rethinking_Deep_Unfolding_Network_in_Illumination_Degradation_Image_Restoration_CVPRF_2026_paper.html
• Code:

UniLDiff: Unlocking the Power of Diffusion Priors for All-in-One Image Restoration

• Paper: https://arxiv.org/abs/2507.23685
• Code:

VideoFusion: A Spatio-Temporal Collaborative Network for Multi-modal Video Fusion and Restoration

• Paper: https://arxiv.org/abs/2503.23359
• Code: https://github.com/Linfeng-Tang/VideoFusion

ZeroIDIR: Zero-Reference Illumination Degradation Image Restoration with Perturbed Consistency Diffusion Models

• Paper: https://arxiv.org/abs/2605.11435
• Code:

7.图像增强(Image Enhancement)

Adapting Large VLMs with Iterative and Manual Instructions for Generative Low-light Enhancement

• Paper: https://openaccess.thecvf.com/content/CVPR2026F/html/Sun_Adapting_Large_VLMs_with_Iterative_and_Manual_Instructions_for_Generative_CVPRF_2026_paper.html
• Code:

Bi-Bridge: Bidirectional Diffusion Bridges for Low-Light Image Enhancement

• Paper: https://arxiv.org/abs/2603.04975
• Code: https://github.com/qq1332427275/Bi-Bridge

BiEvLight: Bi-level Learning of Task-Aware Event Refinement for Low-Light Image Enhancement

• Paper: https://arxiv.org/abs/2603.04975
• Code:

CtrlISP: Rescuing Low-Light RAW Images via Controllable Neural ISP

• Paper: https://openaccess.thecvf.com/content/CVPR2026F/html/Zhang_CtrlISP_Rescuing_Low-Light_RAW_Images_via_Controllable_Neural_ISP_CVPRF_2026_paper.html
• Code:

EIC-LIE: Event-Illumination Collaborative Low-light Image Enhancement

• Paper:
• Code: https://github.com/QUEAHREN/EIC-LIE

Evaluating Low-Light Image Enhancement Across Multiple Intensity Levels

• Paper: https://openaccess.thecvf.com/content/CVPR2026F/html/Pilligua_Evaluating_Low-Light_Image_Enhancement_Across_Multiple_Intensity_Levels_CVPRF_2026_paper.html
• Code:

Event-Illumination Collaborative Low-light Image Enhancement with a High-resolution Real-world Dataset

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Xu_Event-Illumination_Collaborative_Low-light_Image_Enhancement_with_a_High-resolution_Real-world_Dataset_CVPR_2026_paper.html
• Code:

IF-Bench: Benchmarking and Enhancing MLLMs for Infrared Images

• Paper: https://arxiv.org/abs/2512.09663
• Code: https://github.com/casiatao/IF-Bench

Leveraging Multispectral Sensors for Color Correction in Mobile Cameras

• Paper: https://arxiv.org/abs/2512.08441
• Code: https://github.com/LucaCogo/Mobile-Spectral-CC

MR. Illuminate: Zero-Shot Low-Light Image Enhancement with Diffusion Prior

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Cho_MR._Illuminate_Zero-Shot_Low-Light_Image_Enhancement_with_Diffusion_Prior_CVPR_2026_paper.html
• Code:

Multinex: Lightweight Low-Light Image Enhancement via Multi-prior Retinex

• Paper: https://arxiv.org/abs/2604.10359
• Code: https://github.com/albrateanu/multinex

NEC-Diff: Noise-Robust Event-RAW Complementary Diffusion for Seeing Motion in Extreme Darkness

• Paper: https://arxiv.org/abs/2603.20005
• Code: https://github.com/jinghan-xu/NEC-Diff

PrismNet: Semantic-Aware Image Enhancement via Vision Transformer and Zero-Cost Gating

• Paper: https://openaccess.thecvf.com/content/CVPR2026F/html/Zhang_PrismNet_Semantic-Aware_Image_Enhancement_via_Vision_Transformer_and_Zero-Cost_Gating_CVPRF_2026_paper.html
• Code: https://github.com/kuixu/PrismNet

RodNet: Visual Pathway-Inspired Adaptive Sparse Network for Efficient Low-Light Image Enhancement

• Paper: https://openaccess.thecvf.com/content/CVPR2026F/html/Liu_RodNet_Visual_Pathway-Inspired_Adaptive_Sparse_Network_for_Efficient_Low-Light_Image_CVPRF_2026_paper.html
• Code:

SDUIE: Semi-Supervised Diffusion for Underwater Image Enhancement with Quant-Text Dual Control

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Cong_SDUIE_Semi-Supervised_Diffusion_for_Underwater_Image_Enhancement_with_Quant-Text_Dual_CVPR_2026_paper.html
• Code:

Towards Generalized Representations for Low-Light Understanding: When Signal Constancy Meets Semantic Enrichment

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Li_Towards_Generalized_Representations_for_Low-Light_Understanding_When_Signal_Constancy_Meets_CVPR_2026_paper.html
• Code:

VSRELL: A Simple Baseline for Video Super-Resolution and Enhancement in Low-Light environment

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Hui_VSRELL_A_Simple_Baseline_for_Video_Super-Resolution_and_Enhancement_in_CVPR_2026_paper.html
• Code:

8.图像修复(Inpainting)

Blend-Aware Latent Diffusion: Mitigating Stitched Seams in Image Inpainting

• Paper: https://openaccess.thecvf.com/content/CVPR2026F/html/Liu_Blend-Aware_Latent_Diffusion_Mitigating_Stitched_Seams_in_Image_Inpainting_CVPRF_2026_paper.html
• Code:

DreamStereo: Towards Real-Time Stereo Inpainting for HD Videos

• Paper: https://arxiv.org/abs/2604.12270
• Code:

EffectErase: Joint Video Object Removal and Insertion for High-Quality Effect Erasing

• Paper: https://arxiv.org/abs/2603.19224
• Code: https://github.com/FudanCVL/EffectErase

From Inpainting to Layer Decomposition: Repurposing Generative Inpainting Models for Image Layer Decomposition

• Paper: https://arxiv.org/abs/2511.20996
• Code:

GOR-IS: 3D Gaussian Object Removal In the Intrinsic Space

• Paper: https://arxiv.org/abs/2605.00498
• Code: https://github.com/applezyh/GOR-IS

HiFi-Inpaint: Towards High-Fidelity Reference-Based Inpainting for Generating Detail-Preserving Human-Product Images

• Paper: https://arxiv.org/abs/2603.02210
• Code: https://github.com/Correr-Zhou/HiFi-Inpaint

InverFill: One-Step Inversion for Enhanced Few-Step Diffusion Inpainting

• Paper: https://arxiv.org/abs/2603.23463
• Code:

LaRP: Efficient Multi-View Inpainting with Latent Reprojection Priors

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Zhang_LaRP_Efficient_Multi-View_Inpainting_with_Latent_Reprojection_Priors_CVPR_2026_paper.html
• Code:

MAGIC: Few-Shot Mask-Guided Anomaly Inpainting

• Paper: https://arxiv.org/abs/2507.02314
• Code: https://github.com/SpatialAILab/MAGIC-Anomaly-generation

Object-WIPER: Training-Free Object and Associated Effect Removal in Video

• Paper: https://arxiv.org/abs/2601.06391
• Code:

PHAC: Promptable Human Amodal Completion

• Paper: https://arxiv.org/abs/2603.14741
• Code:

Precise Object and Effect Removal with Adaptive Target-Aware Attention

• Paper: https://arxiv.org/abs/2505.22636
• Code: https://github.com/zjx0101/ObjectClear

YOSE: You Only Select Essential Tokens for Efficient DiT-based Video Object Removal

• Paper: https://arxiv.org/abs/2604.27322
• Code: https://github.com/Wucy0519/YOSE-CVPR26

9.高动态范围成像(HDR Imaging)

Beyond8Bits: Full HDR UGC Dataset

• Paper:
• Code: https://github.com/shreshthsaini/Beyond8Bits

ExpoCM: Exposure-Aware One-Step Generative Single-Image HDR Reconstruction

• Paper: https://arxiv.org/abs/2605.02464
• Code:

F²HDR: Two-Stage HDR Video Reconstruction via Flow Adapter and Physical Motion Modeling

• Paper: https://arxiv.org/abs/2603.14920
• Code: https://github.com/wei1895/F2HDR

LRHDR: Learning Representation-enhanced HDR Video Reconstruction

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Liao_LRHDR_Learning_Representation-enhanced_HDR_Video_Reconstruction_CVPR_2026_paper.html
• Code:

Olbedo: An Albedo and Shading Aerial Dataset for Large-Scale Outdoor Environments

• Paper: https://arxiv.org/abs/2602.22025
• Code: https://github.com/GDAOSU/Olbedo

Seeing through Light and Darkness: Sensor-Physics Grounded Deblurring HDR NeRF from Single-Exposure Images and Events

• Paper: https://arxiv.org/abs/2601.15475
• Code: https://github.com/iCVTEAM/See-NeRF

10.图像质量评价(Image Quality Assessment)

A^3: Towards Advertising Aesthetic Assessment

• Paper: https://arxiv.org/abs/2603.24037
• Code: https://github.com/euleryuan/A3-Align

ArtiMuse: Fine-Grained Image Aesthetics Assessment with Joint Scoring and Expert-Level Understanding

• Paper: https://arxiv.org/abs/2507.14533
• Code: https://github.com/thunderbolt215/ArtiMuse

Bridging the Perception Gap in Image Super-Resolution Evaluation

• Paper: https://arxiv.org/abs/2503.13074
• Code:

Fine-grained Image Aesthetic Assessment: Learning Discriminative Scores from Relative Ranks

• Paper: https://arxiv.org/abs/2603.03907
• Code: https://github.com/yzc-ippl/FG-IAA

FluoCLIP: Stain-Aware Focus Quality Assessment in Fluorescence Microscopy

• Paper: https://arxiv.org/abs/2602.23791
• Code: https://github.com/FluoCLIP/FluoCLIP

Generalizable Video Quality Assessment via Weak-to-Strong Learning

• Paper: https://arxiv.org/abs/2505.03631
• Code:

HDR-VLM: HDR-Domain Adaptation of VLMs and Preference-Aligned Quality Assessment for HDR Video Color Grading

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Yuan_HDR-VLM_HDR-Domain_Adaptation_of_VLMs_and_Preference-Aligned_Quality_Assessment_for_CVPR_2026_paper.html
• Code:

Learning Where to Look and How to Judge: Resolution-agnostic Image Quality Assessment with Quality-aware Saliency

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Gedik_Learning_Where_to_Look_and_How_to_Judge_Resolution-agnostic_Image_CVPR_2026_paper.html
• Code:

Life-IQA: Boosting Blind Image Quality Assessment through GCN-enhanced Layer Interaction and MoE-based Feature Decoupling

• Paper: https://arxiv.org/abs/2511.19024
• Code:

MDS-VQA: Model-Informed Data Selection for Video Quality Assessment

• Paper: https://arxiv.org/abs/2603.11525
• Code:

Pioneering Perceptual Video Fluency Assessment: A Novel Task with Benchmark Dataset and Baseline

• Paper: https://arxiv.org/abs/2603.26055
• Code:

Probabilistic Prompt Adaptation for Unified Image Aesthetics and Quality Assessment

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Hara_Probabilistic_Prompt_Adaptation_for_Unified_Image_Aesthetics_and_Quality_Assessment_CVPR_2026_paper.html
• Code:

PR-IQA: Partial-Reference Image Quality Assessment for Diffusion Models

• Paper: https://arxiv.org/abs/2604.04576
• Code: https://github.com/Kakaomacao/PR-IQA

QD-PCQA: Quality-Aware Domain Adaptation for Point Cloud Quality Assessment

• Paper: https://arxiv.org/abs/2603.03726
• Code: https://github.com/huhu-code/QD-PCQA

Rethinking Knowledge Transfer in Image Quality Assessment: A Perceptual Preference Structure Alignment Perspective

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Li_Rethinking_Knowledge_Transfer_in_Image_Quality_Assessment_A_Perceptual_Preference_CVPR_2026_paper.html
• Code:

RL-ScanIQA: Reinforcement-Learned Scanpaths for Blind 360° Image Quality Assessment

• Paper: https://arxiv.org/abs/2603.14297
• Code: https://github.com/wangyuji1/RLScanIQA

rPPG-VQA: A Video Quality Assessment Framework for Unsupervised rPPG Training

• Paper: https://arxiv.org/abs/2604.11156
• Code: https://github.com/Tianyang-Dai/rPPG-VQA

Seeing Beyond 8bits: Subjective and Objective Quality Assessment of HDR-UGC Videos

• Paper: https://arxiv.org/abs/2603.00938
• Code:

VITAL: Vision-Encoder-centered Pre-training for LMMs in Visual Quality Assessment

• Paper: https://arxiv.org/abs/2511.17962
• Code:

11.插帧(Frame Interpolation)

Anchoring and Rescaling Attention for Semantically Coherent Inbetweening

• Paper: https://arxiv.org/abs/2603.17651
• Code: https://github.com/teunchoi/TGI

LDF-VFI: Towards Holistic Modeling for Video Frame Interpolation with Auto-regressive Diffusion Transformers

• Paper: https://arxiv.org/abs/2601.14959
• Code: https://github.com/xypeng9903/LDF-VFI

One-Shot Flow, Any-Time Frame: A Bidirectional Warping Framework for Event-Based Video Frame Interpolation

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Fu_One-Shot_Flow_Any-Time_Frame_A_Bidirectional_Warping_Framework_for_Event-Based_CVPR_2026_paper.html
• Code:

Towards Holistic Modeling for Video Frame Interpolation with Auto-regressive Diffusion Transformers

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Peng_Towards_Holistic_Modeling_for_Video_Frame_Interpolation_with_Auto-regressive_Diffusion_CVPR_2026_paper.html
• Code:

12.视频/图像压缩(Video/Image Compression)

Adaptive Learned Image Compression with Graph Neural Networks

• Paper: https://arxiv.org/abs/2603.25316
• Code: https://github.com/UnoC-727/GLIC

Beyond Pixel Loss: Video-INRs Prefer Perceptual Optimization

• Paper:
• Code:

Block-based Learned Image Compression without Blocking Artifacts

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Kim_Block-based_Learned_Image_Compression_without_Blocking_Artifacts_CVPR_2026_paper.html
• Code:

CADC: Content Adaptive Diffusion-Based Generative Image Compression

• Paper: https://arxiv.org/abs/2602.21591
• Code:

CoD: A Diffusion Foundation Model for Image Compression

• Paper: https://arxiv.org/abs/2511.18706
• Code: https://github.com/microsoft/GenCodec/tree/main/CoD

Differentiable Vector Quantization for Rate-Distortion Optimization of Generative Image Compression

• Paper: https://arxiv.org/abs/2604.10546
• Code: https://github.com/CVL-UESTC/RDVQ

Distributed Image Compression with Multimodal Side Information at Extremely Low Bitrates

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Xu_Distributed_Image_Compression_with_Multimodal_Side_Information_at_Extremely_Low_CVPR_2026_paper.html
• Code:

DiT-IC: Aligned Diffusion Transformer for Efficient Image Compression

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Shi_DiT-IC_Aligned_Diffusion_Transformer_for_Efficient_Image_Compression_CVPR_2026_paper.html
• Code: https://github.com/NJUVISION/DiT-IC

FreqSIC: Frequency-aware Stereo Image Compression with Bi-directional Checkerboard Context Model

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Qin_FreqSIC_Frequency-aware_Stereo_Image_Compression_with_Bi-directional_Checkerboard_Context_Model_CVPR_2026_paper.html
• Code:

Generative Neural Video Compression via Video Diffusion Prior

• Paper: https://arxiv.org/abs/2512.05016
• Code:

Generative Video Compression with One-Dimensional Latent Representation

• Paper: https://arxiv.org/abs/2603.15302
• Code:

Learned Image Compression via Sparse Attention and Adaptive Frequency

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Ma_Learned_Image_Compression_via_Sparse_Attention_and_Adaptive_Frequency_CVPR_2026_paper.html
• Code:

Low-Bitrate Video Compression through Semantic-Conditioned Diffusion

• Paper: https://openaccess.thecvf.com/content/CVPR2026F/html/Wang_Low-Bitrate_Video_Compression_through_Semantic-Conditioned_Diffusion_CVPRF_2026_paper.html
• Code:

MambaSIC: Mamba-based Stereo Image Compression with Bi-directional Multi-reference Entropy Model

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Qin_MambaSIC_Mamba-based_Stereo_Image_Compression_with_Bi-directional_Multi-reference_Entropy_Model_CVPR_2026_paper.html
• Code:

OmniZip: Learning a Unified and Lightweight Lossless Compressor for Multi-Modal Data

• Paper: https://arxiv.org/abs/2602.22286
• Code: https://github.com/adminasmi/OmniZip-CVPR2026

Parallax to Align Them All: An OmniParallax Attention Mechanism for Distributed Multi-View Image Compression

• Paper: https://arxiv.org/abs/2603.03615
• Code:

Perceptual Neural Video Compression with Color Separation and Rank Chain

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Liang_Perceptual_Neural_Video_Compression_with_Color_Separation_and_Rank_Chain_CVPR_2026_paper.html
• Code:

Real-Time Neural Video Compression with Unified Intra and Inter Coding

• Paper: https://arxiv.org/abs/2510.14431
• Code:

Ultra-Fast Neural Video Compression

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Li_Ultra-Fast_Neural_Video_Compression_CVPR_2026_paper.html
• Code:

Ultra-Low Bitrate Perceptual Image Compression with Shallow Encoder

• Paper: https://arxiv.org/abs/2512.12229
• Code: https://github.com/LuizScarlet/AEIC

VLIC: Vision-Language Models As Perceptual Judges for Human-Aligned Image Compression

• Paper: https://arxiv.org/abs/2512.15701
• Code:

What and Where to Adapt: Structure-Semantics Co-Tuning for Machine Vision Compression via Synergistic Adapters

• Paper: https://arxiv.org/abs/2604.10017
• Code: https://github.com/Brock-bit4/S2-CoT

What Matters in Practical Learned Image Compression

• Paper: https://arxiv.org/abs/2605.05148
• Code:

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

14.图像去反光(Image Reflection Removal)

GenSIRR: Rectifying Latent Space for Generative Single-Image Reflection Removal

• Paper:
• Code: https://github.com/lime-j/GenSIRR

GFRRN: Explore the Gaps in Single Image Reflection Removal

• Paper: https://arxiv.org/abs/2602.22695
• Code:

LightRR: A Lightweight Network for Single Image Reflection Removal

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Yin_LightRR_A_Lightweight_Network_for_Single_Image_Reflection_Removal_CVPR_2026_paper.html
• Code:

Polarization State Tracing for Reflection Removal and Color-Consistent Reconstruction

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Wang_Polarization_State_Tracing_for_Reflection_Removal_and_Color-Consistent_Reconstruction_CVPR_2026_paper.html
• Code:

Rectifying Latent Space for Generative Single-Image Reflection Removal

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Li_Rectifying_Latent_Space_for_Generative_Single-Image_Reflection_Removal_CVPR_2026_paper.html
• Code:

Reflection Separation from a Single Image via Joint Latent Diffusion

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Huang_Reflection_Separation_from_a_Single_Image_via_Joint_Latent_Diffusion_CVPR_2026_paper.html
• Code:

ReflexSplit: Single Image Reflection Separation via Layer Fusion-Separation

• Paper: https://www.arxiv.org/abs/2601.17468
• Code: https://github.com/wuw2135/ReflexSplit

15.图像去阴影(Image Shadow Removal)

PhaSR: Generalized Image Shadow Removal with Physically Aligned Priors

• Paper: https://arxiv.org/abs/2601.17470
• Code: https://github.com/ming053l/PhaSR

16.图像上色(Image Colorization)

ColorFLUX: A Structure-Color Decoupling Framework for Old Photo Colorization

• Paper: https://arxiv.org/abs/2603.28162
• Code:

SketchDeco: Training-Free Latent Composition for Precise Sketch Colourisation

• Paper: https://arxiv.org/abs/2405.18716
• Code: https://github.com/CHAITron/sketchdeco-code

Towards High-resolution and Disentangled Reference-based Sketch Colorization

• Paper: https://arxiv.org/abs/2603.05971
• Code: https://github.com/tellurion-kanata/ColorizeDiffusionXL

17.图像和谐化(Image Harmonization)

HarmoniDiff-RS: Training-Free Diffusion Harmonization for Satellite Image Composition

• Paper: https://arxiv.org/abs/2604.19392
• Code: https://github.com/XiaoqiZhuang/HarmoniDiff-RS

HarmoVid: Relightful Video Portrait Harmonization

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Choi_Relightful_Video_Portrait_Harmonization_CVPR_2026_paper.html
• Code:

18.视频稳相(Video Stabilization)

LightStab: Unsupervised Online Video Stabilization with Classical Priors

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Liu_No_Labels_No_Look-Ahead_Unsupervised_Online_Video_Stabilization_with_Classical_CVPR_2026_paper.html
• Code: https://github.com/liutao23/LightStab

No Labels, No Look-Ahead: Unsupervised Online Video Stabilization with Classical Priors

• Paper: https://arxiv.org/abs/2602.23141
• Code:

StabiGS: Video Stabilization through Rendering-Aware Trajectory Optimization in 3DGS-Reconstructed Scenes

• Paper: https://openaccess.thecvf.com/content/CVPR2026F/html/Mabrouk_StabiGS_Video_Stabilization_through_Rendering-Aware_Trajectory_Optimization_in_3DGS-Reconstructed_Scenes_CVPRF_2026_paper.html
• Code:

19.图像融合(Image Fusion)

Beyond Strict Pairing: Arbitrarily Paired Training for High-Performance Infrared and Visible Image Fusion

• Paper: https://arxiv.org/abs/2603.21820
• Code: https://github.com/yanglinDeng/IVIF_unpair

Bridging Human Evaluation to Infrared and Visible Image Fusion

• Paper: https://arxiv.org/abs/2603.03871
• Code:

Customized Fusion: A Closed-Loop Dynamic Network for Adaptive Multi-Task-Aware Infrared-Visible Image Fusion

• Paper: https://arxiv.org/abs/2604.08924
• Code: https://github.com/YR0211/CLDyN

Degradation-Robust Fusion: An Efficient Degradation-Aware Diffusion Framework for Multimodal Image Fusion in Arbitrary Degradation Scenarios

• Paper: https://arxiv.org/abs/2604.08922
• Code:

Fusion in Your Way: Aligning Image Fusion with Heterogeneous Demands via Direct Preference Optimization

• Paper: https://arxiv.org/abs/2605.06049
• Code:

FusionRegister: Every Infrared and Visible Image Fusion Deserves Registration

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Bian_FusionRegister_Every_Infrared_and_Visible_Image_Fusion_Deserves_Registration_CVPR_2026_paper.html
• Code: https://github.com/bociic/FusionRegister

More Than Meets the Eye: A Unified Image Fusion Framework via Semantic-Pixel Entropy Trade-off for Zero-Shot Generalization

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Liu_More_Than_Meets_the_Eye_A_Unified_Image_Fusion_Framework_CVPR_2026_paper.html
• Code:

Multi-Modal Image Fusion via Intervention-Stable Feature Learning

• Paper: https://arxiv.org/abs/2603.23272
• Code:

Missing No More: Dictionary-Guided Cross-Modal Image Fusion under Missing Infrared

• Paper: https://arxiv.org/abs/2603.08018
• Code: https://github.com/harukiv/DCMIF

Neurodynamics-Driven Coupled Neural P Systems for Multi-Focus Image Fusion

• Paper: https://arxiv.org/abs/2509.17704
• Code:

PhyFusion: Physics-Aware Infrared and Visible Image Fusion via Modality-Specific Physical Priors

• Paper: https://openaccess.thecvf.com/content/CVPR2026F/html/Jiang_PhyFusion_Physics-Aware_Infrared_and_Visible_Image_Fusion_via_Modality-Specific_Physical_CVPRF_2026_paper.html
• Code:

ReCoFuse: Ultra-Robust Image Fusion via Restorative Multi-Modal Diffusion Reciprocal Coupling

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Zhang_ReCoFuse_Ultra-Robust_Image_Fusion_via_Restorative_Multi-Modal_Diffusion_Reciprocal_Coupling_CVPR_2026_paper.html
• Code:

RegionFuse: Region-Adaptive Pixel Distribution Learning for Infrared and Visible Image Fusion

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Xia_RegionFuse_Region-Adaptive_Pixel_Distribution_Learning_for_Infrared_and_Visible_Image_CVPR_2026_paper.html
• Code: https://github.com/DarkIceField/RegionFuse

UniFusion: A Unified Image Fusion Framework with Robust Representation and Source-Aware Preservation

• Paper: https://arxiv.org/abs/2603.14214
• Code:

VideoFusion: A Spatio-Temporal Collaborative Network for Multi-modal Video Fusion and Restoration

• Paper: https://arxiv.org/abs/2503.23359
• Code: https://github.com/Linfeng-Tang/VideoFusion

20.其他任务(Others)

3M-TI: High-Quality Mobile Thermal Imaging via Calibration-free Multi-Camera Cross-Modal Diffusion

• Paper: https://arxiv.org/abs/2511.19117
• Code: https://github.com/work-submit/3MTI

AceTone: Bridging Words and Colors for Conditional Image Grading

• Paper: https://arxiv.org/abs/2604.00530
• Code: https://github.com/martian422/AceTone

Continuous Exposure-Time Modeling for Realistic Atmospheric Turbulence Synthesis

• Paper: https://arxiv.org/abs/2603.01398
• Code: https://github.com/Jun-Wei-Zeng/ET-Turb

Cross-Scale Pansharpening via ScaleFormer and the PanScale Benchmark

• Paper: https://arxiv.org/abs/2603.00543
• Code: https://github.com/caoke-963/ScaleFormer

CRFT: Consistent-Recurrent Feature Flow Transformer for Cross-Modal Image Registration

• Paper: https://arxiv.org/abs/2604.05689
• Code: https://github.com/NEU-Liuxuecong/CRFT

D2Dewarp: Dual Dimensions Geometric Representation Learning Based Document Image Dewarping

• Paper: https://arxiv.org/abs/2507.08492
• Code: https://github.com/xiaomore/D2Dewarp

Dark3R: Learning Structure from Motion in the Dark

• Paper: https://arxiv.org/abs/2603.05330
• Code: https://github.com/andrewyguo/Dark3R

Disentangle-then-Align: Non-Iterative Hybrid Multimodal Image Registration via Cross-Scale Feature Disentanglement

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Zhang_Disentangle-then-Align_Non-Iterative_Hybrid_Multimodal_Image_Registration_via_Cross-Scale_Feature_Disentanglement_CVPR_2026_paper.html
• Code: https://github.com/Chunlei0913/HRNet

Elucidating the Design Space of Arbitrary-Noise-Based Diffusion Models

• Paper: https://arxiv.org/abs/2507.18534
• Code: https://github.com/PerceptionComputingLab/EDA

Exploring Spatiotemporal Feature Propagation for Video-Level Compressive Spectral Reconstruction: Dataset, Model and Benchmark

• Paper: https://arxiv.org/abs/2603.00611
• Code: https://github.com/nju-cite/DynaSpec

FastGaMer: Efficient GainMap Learning for Practical Inverse Tone Mapping

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Guan_FastGaMer_Efficient_GainMap_Learning_for_Practical_Inverse_Tone_Mapping_CVPR_2026_paper.html
• Code:

Fast Kernel-Space Diffusion for Remote Sensing Pansharpening

• Paper: https://openaccess.thecvf.com/content/CVPR2026F/html/Jin_Fast_Kernel-Space_Diffusion_for_Remote_Sensing_Pansharpening_CVPRF_2026_paper.html
• Code:

HFR and HDR Video from Multi-Attenuated Spikes Using a Rapidly Rotating SpokeND Filter

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Chang_HFR_and_HDR_Video_from_Multi-Attenuated_Spikes_Using_a_Rapidly_CVPR_2026_paper.html
• Code:

High-Quality and Efficient Turbulence Mitigation with Events

• Paper: https://arxiv.org/abs/2603.20708
• Code: https://github.com/Xavier667/EHETM

InstantRetouch: Efficient and High-Fidelity Instruction-Guided Image Retouching with Bilateral Space

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Wu_InstantRetouch_Efficient_and_High-Fidelity_Instruction-Guided_Image_Retouching_with_Bilateral_Space_CVPR_2026_paper.html
• Code: https://github.com/OpenImagingLab/InstantRetouch

It Takes Two: A Duet of Periodicity and Directionality for Burst Flicker Removal

• Paper: https://arxiv.org/abs/2603.22794
• Code: https://github.com/qulishen/Flickerformer

JarvisEvo: Towards a Self-Evolving Photo Editing Agent with Synergistic Editor-Evaluator Optimization

• Paper: https://arxiv.org/abs/2511.23002
• Code: https://github.com/LYL1015/JarvisEvo

Language-Guided One-Step Diffusion Model for Nighttime Flare Removal

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Ning_Language-Guided_One-Step_Diffusion_Model_for_Nighttime_Flare_Removal_CVPR_2026_paper.html
• Code:

Learning Latent Transmission and Glare Maps for Lens Veiling Glare Removal

• Paper: https://arxiv.org/abs/2511.17353
• Code: https://github.com/XiaolongQian/DeVeiler

LRDUN: A Low-Rank Deep Unfolding Network for Efficient Spectral Compressive Imaging

• Paper: https://arxiv.org/abs/2511.18513
• Code: https://github.com/huang-he99/LRDUN

MERIT: Multi-domain Efficient RAW Image Translation

• Paper: https://arxiv.org/abs/2603.20836
• Code:

MTRWKV: Multigrain-aware Semantic Prototype Scanning and Tri-Token Prompt Learning Embraced High-Order RWKV for Pan-Sharpening

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Li_Multigrain-aware_Semantic_Prototype_Scanning_and_Tri-Token_Prompt_Learning_Embraced_High-Order_CVPR_2026_paper.html
• Code: https://github.com/L1junfeng/MTRWKV

Multigrain-aware Semantic Prototype Scanning and Tri-Token Prompt Learning Embraced High-Order RWKV for Pan-Sharpening

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Li_Multigrain-aware_Semantic_Prototype_Scanning_and_Tri-Token_Prompt_Learning_Embraced_High-Order_CVPR_2026_paper.html
• Code:

POS-ISP: Pipeline Optimization at the Sequence Level for Task-aware ISP

• Paper: https://arxiv.org/abs/2604.06938
• Code: https://github.com/w1jyun/POS-ISP

PromptStereo: Zero-Shot Stereo Matching via Structure and Motion Prompts

• Paper: https://arxiv.org/abs/2603.01650
• Code:

Regulating Rather than Constraining: Adaptive Guidance for Complex Spectral Reconstruction in Pansharpening

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Wen_Regulating_Rather_than_Constraining_Adaptive_Guidance_for_Complex_Spectral_Reconstruction_CVPR_2026_paper.html
• Code:

RetouchIQ: MLLM Agents for Instruction-Based Image Retouching with Generalist Reward

• Paper: https://arxiv.org/abs/2602.17558
• Code:

Reparameterized Tensor Ring Functional Decomposition for Multi-Dimensional Data Recovery

• Paper: https://arxiv.org/abs/2603.01034
• Code: https://github.com/YangyangXu2002/RepTRFD

Seeing Through Blur: Tackling Defocus in Spike-Based Imaging

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Ma_Seeing_Through_Blur_Tackling_Defocus_in_Spike-Based_Imaging_CVPR_2026_paper.html
• Code:

Spatial-Spectral Residuals Informed Diffusion Neural Operator for Pan-sharpening

• Paper: https://openaccess.thecvf.com/content/CVPR2026/html/Huang_Spatial-Spectral_Residuals_Informed_Diffusion_Neural_Operator_for_Pan-sharpening_CVPR_2026_paper.html
• Code:

SpiralDiff: Spiral Diffusion with LoRA for RGB-to-RAW Conversion Across Cameras

• Paper: https://arxiv.org/abs/2603.14885
• Code: https://github.com/Chuancy-TJU/SpiralDiff

Stability and Non-Local Modeling in Hybrid Convolution-Transformer Networks for Snapshot Hyperspectral Reconstruction

• Paper: https://openaccess.thecvf.com/content/CVPR2026F/html/Han_Stability_and_Non-Local_Modeling_in_Hybrid_Convolution-Transformer_Networks_for_Snapshot_CVPRF_2026_paper.html
• Code:

Towards Photorealistic and Efficient Bokeh Rendering via Diffusion Framework

• Paper: https://arxiv.org/abs/2605.07429
• Code:

Towards Universal Computational Aberration Correction in Photographic Cameras: A Comprehensive Benchmark Analysis

• Paper: https://arxiv.org/abs/2603.12083
• Code: https://github.com/XiaolongQian/UniCAC

UnReflectAnything: RGB-Only Highlight Removal by Rendering Synthetic Specular Supervision

• Paper: https://arxiv.org/abs/2512.09583
• Code: https://github.com/alberto-rota/UnReflectAnything

White-Balance First, Adjust Later: Cross-Camera Color Constancy via Vision-Language Evaluation

• Paper: https://arxiv.org/abs/2605.19613v1
• Code:

持续更新~