CVPR2021.md

Awesome-CVPR2021-Low-Level-Vision

整理汇总下今年CVPR图像重建(Image Reconstruction)/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。大家如果觉得有帮助，欢迎star~~

参考或转载请注明出处

CVPR2021官网：http://cvpr2021.thecvf.com

CVPR完整论文列表：https://openaccess.thecvf.com/CVPR2021

开会时间：2021年6月19日-6月25日

论文接收公布时间：2021年2月28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.图像去摩尔纹（Image Demoireing）
• 9.图像修复（Inpainting）
• 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.其他多任务

1.超分辨率（Super-Resolution）

Unsupervised Degradation Representation Learning for Blind Super-Resolution

• Paper：https://arxiv.org/abs/2104.00416
• Code：https://github.com/LongguangWang/DASR

Data-Free Knowledge Distillation For Image Super-Resolution

• Paper：https://openaccess.thecvf.com/content/CVPR2021/papers/Zhang_Data-Free_Knowledge_Distillation_for_Image_Super-Resolution_CVPR_2021_paper.pdf

AdderSR: Towards Energy Efficient Image Super-Resolution

• Paper：https://arxiv.org/abs/2009.08891
• Code：

Exploring Sparsity in Image Super-Resolution for Efficient Inference

• Paper：https://arxiv.org/abs/2006.09603
• Code：https://github.com/LongguangWang/SMSR

ClassSR: A General Framework to Accelerate Super-Resolution Networks by Data Characteristic

• Paper：https://arxiv.org/abs/2103.04039
• Code：https://github.com/Xiangtaokong/ClassSR

Cross-MPI: Cross-scale Stereo for Image Super-Resolution using Multiplane Images

• Paper：https://arxiv.org/abs/2011.14631
• Code：
• Homepage：http://www.liuyebin.com/crossMPI/crossMPI.html
• Analysis：CVPR 2021，Cross-MPI以底层场景结构为线索的端到端网络，在大分辨率（x8）差距下也可完成高保真的超分辨率

LAU-Net: Latitude Adaptive Upscaling Network for Omnidirectional Image Super-resolution

• Paper：https://openaccess.thecvf.com/content/CVPR2021/html/Deng_LAU-Net_Latitude_Adaptive_Upscaling_Network_for_Omnidirectional_Image_Super-Resolution_CVPR_2021_paper.html
• Code：https://github.com/wangh-allen/LAU-Net

Learning Continuous Image Representation with Local Implicit Image Function

• Paper：https://arxiv.org/abs/2012.09161
• Code：https://github.com/yinboc/liif
• Homepage：https://yinboc.github.io/liif/

Temporal Modulation Network for Controllable Space-Time Video Super-Resolution

• Paper：https://arxiv.org/abs/2104.10642
• Code：https://github.com/CS-GangXu/TMNet

Robust Reference-based Super-Resolution via C²-Matching

• Paper：https://openaccess.thecvf.com/content/CVPR2021/html/Jiang_Robust_Reference-Based_Super-Resolution_via_C2-Matching_CVPR_2021_paper.html
• Code：https://github.com/yumingj/C2-Matching

GLEAN: Generative Latent Bank for Large-Factor Image Super-Resolution

• Paper：https://ckkelvinchan.github.io/papers/glean.pdf
• Code： https://github.com/ckkelvinchan/GLEAN
• Homepage：https://ckkelvinchan.github.io/projects/GLEAN/

BasicVSR: The Search for Essential Components in Video Super-Resolution and Beyond

• Paper：https://arxiv.org/abs/2012.02181
• Code：https://github.com/ckkelvinchan/BasicVSR-IconVSR
• Homepage：https://ckkelvinchan.github.io/projects/BasicVSR/

Video Rescaling Networks with Joint Optimization Strategies for Downscaling and Upscaling

• Paper：https://arxiv.org/abs/2103.14858
• Code：https://github.com/ding3820/MIMO-VRN
• Homepage：https://ding3820.github.io/MIMO-VRN/

MASA-SR: Matching Acceleration and Spatial Adaptation for Reference-Based Image Super-Resolution

• Paper：https://jiaya.me/papers/masasr_cvpr21.pdf
• Code：https://github.com/Jia-Research-Lab/MASA-SR

Flow-based Kernel Prior with Application to Blind Super-Resolution

• Paper：https://arxiv.org/pdf/2103.15977.pdf
• Code：https://github.com/JingyunLiang/FKP

Interpreting Super-Resolution Networks with Local Attribution Maps

• Paper:https://arxiv.org/abs/2011.11036v1
• Homepage：https://x-lowlevel-vision.github.io/lam.html
• Analysis：https://zhuanlan.zhihu.com/p/363139999

SRWarp: Generalized Image Super-Resolution under Arbitrary Transformation

• Paper：https://arxiv.org/abs/2104.10325
• Code：https://github.com/sanghyun-son/srwarp

KOALAnet: Blind Super-Resolution using Kernel-Oriented Adaptive Local Adjustment

• Paper：https://arxiv.org/abs/2012.08103
• Code：https://github.com/hjSim/KOALAnet

Towards Fast and Accurate Real-World Depth Super-Resolution: Benchmark Dataset and Baseline

• Paper：https://openaccess.thecvf.com/content/CVPR2021/html/He_Towards_Fast_and_Accurate_Real-World_Depth_Super-Resolution_Benchmark_Dataset_and_CVPR_2021_paper.html

Tackling the Ill-Posedness of Super-Resolution Through Adaptive Target Generation

• Paper：https://openaccess.thecvf.com/content/CVPR2021/html/Jo_Tackling_the_Ill-Posedness_of_Super-Resolution_Through_Adaptive_Target_Generation_CVPR_2021_paper.html
• Code：https://github.com/yhjo09/AdaTarget

Image Super-Resolution With Non-Local Sparse Attention

• Paper：https://openaccess.thecvf.com/content/CVPR2021/html/Mei_Image_Super-Resolution_With_Non-Local_Sparse_Attention_CVPR_2021_paper.html
• Code：https://github.com/HarukiYqM/Non-Local-Sparse-Attention

Unsupervised Real-World Image Super Resolution via Domain-Distance Aware Training

• Paper：https://arxiv.org/abs/2004.01178
• Code：https://github.com/ShuhangGu/DASR

Single Pair Cross-Modality Super Resolution

• Paper：https://openaccess.thecvf.com/content/CVPR2021/html/Shacht_Single_Pair_Cross-Modality_Super_Resolution_CVPR_2021_paper.html

Learning Scene Structure Guidance via Cross-Task Knowledge Transfer for Single Depth Super-Resolution

• Paper：https://openaccess.thecvf.com/content/CVPR2021/html/Sun_Learning_Scene_Structure_Guidance_via_Cross-Task_Knowledge_Transfer_for_Single_CVPR_2021_paper.html
• Code：https://github.com/Sunbaoli/dsr-distillation

Deep Burst Super-Resolution

• Paper：https://openaccess.thecvf.com/content/CVPR2021/html/Bhat_Deep_Burst_Super-Resolution_CVPR_2021_paper.html
• Code：https://github.com/goutamgmb/NTIRE21_BURSTSR

Learning the Non-Differentiable Optimization for Blind Super-Resolution

• Paper：https://openaccess.thecvf.com/content/CVPR2021/html/Hui_Learning_the_Non-Differentiable_Optimization_for_Blind_Super-Resolution_CVPR_2021_paper.html

Light Field Super-Resolution With Zero-Shot Learning

• Paper：https://openaccess.thecvf.com/content/CVPR2021/html/Cheng_Light_Field_Super-Resolution_With_Zero-Shot_Learning_CVPR_2021_paper.html

Space-Time Distillation for Video Super-Resolution

• Paper：https://openaccess.thecvf.com/content/CVPR2021/html/Xiao_Space-Time_Distillation_for_Video_Super-Resolution_CVPR_2021_paper.html

EventZoom: Learning To Denoise and Super Resolve Neuromorphic Events

• Paper：https://openaccess.thecvf.com/content/CVPR2021/html/Duan_EventZoom_Learning_To_Denoise_and_Super_Resolve_Neuromorphic_Events_CVPR_2021_paper.html

MR Image Super-Resolution With Squeeze and Excitation Reasoning Attention Network

• Paper：https://openaccess.thecvf.com/content/CVPR2021/html/Zhang_MR_Image_Super-Resolution_With_Squeeze_and_Excitation_Reasoning_Attention_Network_CVPR_2021_paper.html

Turning Frequency to Resolution: Video Super-Resolution via Event Cameras

• Paper：https://openaccess.thecvf.com/content/CVPR2021/html/Jing_Turning_Frequency_to_Resolution_Video_Super-Resolution_via_Event_Cameras_CVPR_2021_paper.html

Fast Bayesian Uncertainty Estimation and Reduction of Batch Normalized Single Image Super-Resolution Network

• Paper：https://openaccess.thecvf.com/content/CVPR2021/html/Kar_Fast_Bayesian_Uncertainty_Estimation_and_Reduction_of_Batch_Normalized_Single_CVPR_2021_paper.html
• Code：https://github.com/aupendu/sr-uncertainty

Practical Single-Image Super-Resolution Using Look-Up Table

• Paper：https://openaccess.thecvf.com/content/CVPR2021/html/Jo_Practical_Single-Image_Super-Resolution_Using_Look-Up_Table_CVPR_2021_paper.html
• Code：https://github.com/yhjo09/SR-LUT

Interpreting Super-Resolution Networks With Local Attribution Maps

• Paper：https://openaccess.thecvf.com/content/CVPR2021/html/Gu_Interpreting_Super-Resolution_Networks_With_Local_Attribution_Maps_CVPR_2021_paper.html

Scene Text Telescope: Text-Focused Scene Image Super-Resolution

• Paper：https://openaccess.thecvf.com/content/CVPR2021/html/Chen_Scene_Text_Telescope_Text-Focused_Scene_Image_Super-Resolution_CVPR_2021_paper.html
• Code：https://github.com/FudanVI/FudanOCR

GAN Prior Embedded Network for Blind Face Restoration in the Wild

• Paper：https://arxiv.org/abs/2105.06070
• Code：https://github.com/yangxy/GPEN

2.图像去雨（Image Deraining）

Removing Raindrops and Rain Streaks in One Go

• Paper：https://www.researchgate.net/publication/350755019_Removing_Raindrops_and_Rain_Streaks_in_One_Go

From Rain Generation to Rain Removal

• Paper：https://arxiv.org/abs/2008.03580
• Code：https://github.com/hongwang01/VRGNet

Semi-Supervised Video Deraining Embedded with Dynamical Rain Generator

• Paper：https://arxiv.org/abs/2103.07939
• Code：https://github.com/zsyOAOA/S2VD

Closing the Loop: Joint Rain Generation and Removal via Disentangled Image Translation

• Paper：https://arxiv.org/abs/2103.13660
• Code：https://github.com/guyii54/JRGR

3.图像去雾（Image Dehazing）

Learning to Restore Hazy Video: A New Real-World Dataset and A New Method

ContrastiveLearning for Compact Single Image Dehazing

• Paper：https://arxiv.org/abs/2104.09367
• Code：https://github.com/GlassyWu/AECR-Net

4.去模糊（Deblurring）

DeFMO: Deblurring and Shape Recovery of Fast Moving Objects

• Paper：https://arxiv.org/abs/2012.00595
• Code：https://github.com/rozumden/DeFMO

ARVo: Learning All-Range Volumetric Correspondence for Video Deblurring

• Paper：https://arxiv.org/abs/2103.04260

Towards Rolling Shutter Correction and Deblurring in Dynamic Scenes

• Paper：https://arxiv.org/abs/2104.01601
• Code：https://github.com/zzh-tech/RSCD

Explore Image Deblurring via Blur Kernel Space

• Paper：https://arxiv.org/abs/2104.00317

Digital Gimbal: End-to-end Deep Image Stabilization with Learnable Exposure Times

• Paper：https://arxiv.org/abs/2012.04515

5.去噪（Denoising）

Neighbor2Neighbor: Self-Supervised Denoising from Single Noisy Images

• Paper：https://arxiv.org/abs/2101.02824
• Code：https://github.com/TaoHuang2018/Neighbor2Neighbor

NBNet: Noise Basis Learning for Image Denoising with Subspace Projection

• Paper：https://arxiv.org/abs/2012.15028
• Code：https://github.com/megvii-research/NBNet

Beyond Joint Demosaicking and Denoising

• Paper：https://arxiv.org/abs/2104.09398
• Code：https://github.com/sharif-apu/BJDD_CVPR21

Deep Denoising of Flash and No-Flash Pairs for Photography in Low-Light Environments

• Paper：https://arxiv.org/abs/2012.05116
• Code：
• Homepage:https://www.cse.wustl.edu/~zhihao.xia/deepfnf/

Invertible Denoising Network: A Light Solution for Real Noise Removal

• Paper：https://arxiv.org/abs/2104.10546v1
• Code：https://github.com/Yang-Liu1082/InvDN

FBI-Denoiser: Fast Blind Image Denoiser for Poisson-Gaussian Noise

• Paper：https://arxiv.org/abs/2105.10967
• Code：https://github.com/csm9493/FBI-Denoiser

6.图像恢复（Image Restoration）

Multi-Stage Progressive Image Restoration

• Paper：https://arxiv.org/abs/2102.02808
• Code：https://github.com/swz30/MPRNet
• Analysis：

CT Film Recovery via Disentangling Geometric Deformation and Illumination Variation: Simulated Datasets and Deep Models

• Paper：https://arxiv.org/abs/2012.09491
• Code：https://github.com/transcendentsky/Film-Recovery

Restoring Extremely Dark Images in Real Time

• Code：https://github.com/MohitLamba94/Restoring-Extremely-Dark-Images-In-Real-Time

Dual Pixel Exploration: Simultaneous Depth Estimation and Image Restoration

• Paper：https://arxiv.org/abs/2012.00301
• Code：https://github.com/panpanfei/Dual-Pixel-Exploration-Simultaneous-Depth-Estimation-and-Image-Restoration

Progressive Semantic-Aware Style Transformation for Blind Face Restoration

• Paper：https://arxiv.org/abs/2009.08709
• Code：https://github.com/chaofengc/PSFRGAN

Towards Real-World Blind Face Restoration with Generative Facial Prior

• Paper：https://arxiv.org/abs/2101.04061

GFP-GAN: Towards Real-World Blind Face Restoration with Generative Facial Prior

• Paper：https://arxiv.org/abs/2101.04061
• Code：https://github.com/TencentARC/GFPGAN

7.图像增强（Image Enhancement）

Auto-Exposure Fusion for Single-Image Shadow Removal

• Paper：https://arxiv.org/abs/2103.01255
• Code：https://github.com/tsingqguo/exposure-fusion-shadow-removal

Learning Multi-Scale Photo Exposure Correction

• Paper：https://arxiv.org/abs/2003.11596
• Code：https://github.com/mahmoudnafifi/Exposure_Correction

Robust Reflection Removal with Reflection-free Flash-only Cues

• Paper：https://arxiv.org/abs/2103.04273
• Code：https://github.com/ChenyangLEI/flash-reflection-removal

Learning Temporal Consistency for Low Light Video Enhancement from Single Images

• Code：https://github.com/zkawfanx/StableLLVE

Removing Diffraction Image Artifacts in Under-Display Camera via Dynamic Skip Connection Network

• Paper：https://arxiv.org/abs/2104.09556
• Code：https://github.com/jnjaby/DISCNet

From Shadow Generation to Shadow Removal

• Paper：https://arxiv.org/abs/2103.12997
• Code：https://github.com/hhqweasd/G2R-ShadowNet

PPR10K: A Large-Scale Portrait Photo Retouching Dataset with Human-Region Mask and Group-Level Consistency

• Paper：http://www4.comp.polyu.edu.hk/~cslzhang/paper/PPR10K-cvpr21-paper.pdf
• Code：https://github.com/csjliang/PPR10K

8.图像去摩尔纹（Image Demoireing）

9.图像修复（Inpainting）

PD-GAN:Probabilistic Diverse GAN for Image Inpainting

• Paper：https://arxiv.org/abs/2105.02201
• Code：https://github.com/KumapowerLIU/PD-GAN

Generating Diverse Structure for Image Inpainting with Hierarchical VQ-VAE

• Paper：https://arxiv.org/abs/2103.10022
• Code：https://github.com/USTC-JialunPeng/Diverse-Structure-Inpainting

Image Inpainting with External-internal Learning and Monochromic Bottleneck

• Paper：https://www.cqf.io/papers/Image_Inpainting_Monochromic_Bottleneck_CVPR2021.pdf
• Code：https://github.com/Tengfei-Wang/external-internal-inpainting
• Homepage：https://tengfei-wang.github.io/EII/index.html

Progressive Temporal Feature Alignment Network for Video Inpainting

• Paper：https://arxiv.org/abs/2104.03507v1
• Code：https://github.com/MaureenZOU/TSAM

TransFill: Reference-guided Image Inpainting by Merging Multiple Color and Spatial Transformations

• Paper：https://arxiv.org/abs/2103.15982
• Code：https://github.com/yzhouas/TransFill-Reference-Inpainting

GFP-GAN: Towards Real-World Blind Face Restoration with Generative Facial Prior

• Paper：https://arxiv.org/abs/2101.04061
• Homepage：https://xinntao.github.io/projects/gfpgan

10.图像质量评价（Image Quality Assessment）

SDD-FIQA:Unsupervised Face Image Quality Assessment with Similarity DistributionDistance

• Paper：https://arxiv.org/abs/2103.05977
• Code：https://github.com/Slinene/SDD-FIQA

Neural Side-by-Side: Predicting Human Preferences for No-Reference Super-Resolution Evaluation

• Paper：https://openaccess.thecvf.com/content/CVPR2021/html/Khrulkov_Neural_Side-by-Side_Predicting_Human_Preferences_for_No-Reference_Super-Resolution_Evaluation_CVPR_2021_paper.html

11.插帧（Frame Interpolation）

FLAVR: Flow-Agnostic Video Representations for Fast Frame Interpolation

• Paper：https://arxiv.org/abs/2012.08512
• Code：https://tarun005.github.io/FLAVR/Code
• Homepage：https://tarun005.github.io/FLAVR/

CDFI: Compression-driven Network Design for Frame Interpolation

• Paper：https://arxiv.org/abs/2103.10559
• Code:https://github.com/tding1/Compression-Driven-Frame-Interpolation

DeFMO: Deblurring and Shape Recovery of Fast Moving Objects

• Paper：https://arxiv.org/abs/2012.00595
• Code：https://github.com/rozumden/DeFMO

Deep Animation Video Interpolation in the Wild

• Paper：https://arxiv.org/abs/2104.02495
• Code：https://github.com/lisiyao21/AnimeInterp

12.视频/图像压缩（Video/Image Compression）

MetaSCI: Scalable and Adaptive Reconstruction for Video Compressive Sensing

• Paper：https://arxiv.org/abs/2103.01786
• Code：https://github.com/xyvirtualgroup/MetaSCI-CVPR2021

FVC: A New Framework towards Deep Video Compression in Feature Space

• Paper：https://arxiv.org/abs/2105.09600

Deep Learning in Latent Space for Video Prediction and Compression

• Paper：https://openaccess.thecvf.com/content/CVPR2021/html/Liu_Deep_Learning_in_Latent_Space_for_Video_Prediction_and_Compression_CVPR_2021_paper.html

Deep Perceptual Preprocessing for Video Coding

• Paper：https://openaccess.thecvf.com/content/CVPR2021/html/Chadha_Deep_Perceptual_Preprocessing_for_Video_Coding_CVPR_2021_paper.html

Checkerboard Context Model for Efficient Learned Image Compression

• Paper: https://arxiv.org/abs/2103.15306

Slimmable Compressive Autoencoders for Practical Neural Image Compression

• Paper: https://arxiv.org/abs/2103.15726
• Code：https://github.com/FireFYF/SlimCAE

Attention-guided Image Compression by Deep Reconstruction of Compressive Sensed Saliency Skeleton

• Paper: https://arxiv.org/abs/2103.15368
• Code：https://github.com/xzhang9308/AGDL

Deep Homography for Efficient Stereo Image Compression

• Paper: http://buaamc2.net/pdf/cvpr21hesic.pdf

How To Exploit the Transferability of Learned Image Compression to Conventional Codecs

• Paper: https://arxiv.org/abs/2012.01874

Learning Scalable lY=-Constrained Near-Lossless Image Compression via Joint Lossy Image and Residual Compression

• Paper: https://openaccess.thecvf.com/content/CVPR2021/html/Bai_Learning_Scalable_lY-Constrained_Near-Lossless_Image_Compression_via_Joint_Lossy_Image_CVPR_2021_paper.html

Asymmetric Gained Deep Image Compression With Continuous Rate Adaptation

• Paper: https://openaccess.thecvf.com/content/CVPR2021/html/Cui_Asymmetric_Gained_Deep_Image_Compression_With_Continuous_Rate_Adaptation_CVPR_2021_paper.html

13.其他多任务

Pre-Trained Image Processing Transformer

• Paper：https://arxiv.org/abs/2012.00364
• Code：https://github.com/huawei-noah/Pretrained-IPT
• Analysis：CVPR 2021 | Transformer进军low-level视觉！北大华为等提出预训练模型IPT

Invertible Image Signal Processing

• Paper：https://arxiv.org/abs/2103.15061
• Code：https://github.com/yzxing87/Invertible-ISP

End-to-End Learning for Joint Image Demosaicing, Denoising and Super-Resolution

• Paper：https://openaccess.thecvf.com/content/CVPR2021/html/Xing_End-to-End_Learning_for_Joint_Image_Demosaicing_Denoising_and_Super-Resolution_CVPR_2021_paper.html

持续更新~

参考

[1] CVPR 2021 结果出炉！最新71篇CVPR'21论文汇总（更新中）

[2] CVPR2021最新信息及已接收论文/代码(持续更新)

[3] 15分钟看完：悉尼科技大学入选 CVPR 2021 的 13 篇论文，都研究什么？

[4] CVPR 2021放榜，腾讯优图20篇论文都在这里了

相关Low-Level-Vision整理

• Awesome-CVPR2020-Low-Level-Vision
• Awesome-ECCV2020-Low-Level-Vision