Awesome-CVPR2020-Low-Level-Vision

A Collection of Papers and Codes for CVPR2020 Low Level Vision or Image Reconstruction

整理汇总了下今年CVPR图像重建(Image Reconstruction)/底层视觉(Low-Level Vision)相关的一些论文，包括超分辨率，图像恢复，去雨，去雾，去模糊，去噪等方向。大家如果觉得有帮助，欢迎star~~

• CVPR2020的所有论文：http://openaccess.thecvf.com/CVPR2020.py
• CVPR2020Workshops：http://openaccess.thecvf.com/CVPR2020_workshops/menu.py

【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.其他

1.超分辨率（Super-Resolution）

图像超分辨率

PULSE: Self-Supervised Photo Upsampling via Latent Space Exploration of Generative Models

• Paper：https://arxiv.org/abs/2003.03808
• Code：https://github.com/adamian98/pulse
• Analysis：杜克大学提出 AI 算法，拯救渣画质马赛克秒变高清

Closed-Loop Matters: Dual Regression Networks for Single Image Super-Resolution

• Paper：https://arxiv.org/abs/2003.07018
• Code：https://github.com/guoyongcs/DRN
• Analysis：CVPR2020丨DRN：用于单图像超分辨率的对偶回归网络

EventSR: From Asynchronous Events to Image Reconstruction, Restoration, and Super-Resolution via End-to-End Adversarial Learning

• Paper：http://openaccess.thecvf.com/content_CVPR_2020/papers/Wang_EventSR_From_Asynchronous_Events_to_Image_Reconstruction_Restoration_and_Super-Resolution_CVPR_2020_paper.pdf
• Video ：https://www.youtube.com/watch?v=OShS_MwHecs
• Dataset： https://github.com/wl082013/ESIM_dataset

Unpaired Image Super-Resolution Using Pseudo-Supervision

• Paper：https://arxiv.org/abs/2002.11397?context=eess
• Analysis：#每日五分钟一读#Image Super-Resolution

Correction Filter for Single Image Super-Resolution: Robustifying Off-the-Shelf Deep Super-Resolvers

• Paper：https://arxiv.org/abs/1912.00157
• Code：https://github.com/shadyabh/Correction-Filter
• Analysis：Correction Filter for Single Image Super-Resolution阅读笔记（CVPR2020）

Residual Feature Aggregation Network for Image Super-Resolution

• Paper：http://openaccess.thecvf.com/content_CVPR_2020/papers/Liu_Residual_Feature_Aggregation_Network_for_Image_Super-Resolution_CVPR_2020_paper.pdf
• Code：https://github.com/njulj/RFANet
• Analysis：超越RCAN，图像超分又一峰：RFANet

Deep Unfolding Network for Image Super-Resolution

• Paper：https://arxiv.org/abs/2003.10428
• Code：https://github.com/cszn/USRNet
• Analysis：CVPR2020：USRNet

Image Super-Resolution With Cross-Scale Non-Local Attention and Exhaustive Self-Exemplars Mining

• Paper：https://arxiv.org/abs/2006.01424
• Code：https://github.com/SHI-Labs/Cross-Scale-Non-Local-Attention
• Analysis：CVPR2020|跨尺度非局部注意力模型改进图像超分辨率，Code开源

Learning Texture Transformer Network for Image Super-Resolution

• Paper：https://arxiv.org/abs/2006.04139
• Code：https://github.com/FuzhiYang/TTSR
• Analysis：CVPR 2020丨图像超清化+老照片修复技术，拯救你所有的模糊、破损照片

Robust Reference-Based Super-Resolution With Similarity-Aware Deformable Convolution

• Paper：http://openaccess.thecvf.com/content_CVPR_2020/html/Shim_Robust_Reference-Based_Super-Resolution_With_Similarity-Aware_Deformable_Convolution_CVPR_2020_paper.html
• Analysis：8/19 (CVPR2020) Robust Reference-based Super-Resolution with Similarity-Aware Deformable Convolution

Structure-Preserving Super Resolution with Gradient Guidance

• Paper：https://arxiv.org/abs/2003.13063
• Code：https://github.com/Maclory/Deep-Iterative-Collaboration
• Analysis：CVPR2020丨SPSR：基于梯度指导的结构保留超分辨率方法

Unified Dynamic Convolutional Network for Super-Resolution With Variational Degradations

• Paper：https://arxiv.org/abs/2004.06965
• Analysis：UDVD：适用于可变降质类型的通用图像超分，附参考Code

Perceptual Extreme Super Resolution Network with Receptive Field Block

• Paper：https://arxiv.org/abs/2005.12597
• Analysis：NTIRE2020冠军方案RFB-ESRGAN：带感受野模块的超分网络
• Remarks：NTIRE2020极限超分冠军方案RFB-ESRGAN；Workshops

Real-World Super-Resolution via Kernel Estimation and Noise Injection

• Paper：http://openaccess.thecvf.com/content_CVPRW_2020/html/w31/Ji_Real-World_Super-Resolution_via_Kernel_Estimation_and_Noise_Injection_CVPRW_2020_paper.html
• Code：https://github.com/jixiaozhong/RealSR
• Remarks：NTIRE2020-RWSR超分双赛道冠军方案；Workshops

Investigating Loss Functions for Extreme Super-Resolution

• Paper：http://openaccess.thecvf.com/content_CVPRW_2020/papers/w31/Jo_Investigating_Loss_Functions_for_Extreme_Super-Resolution_CVPRW_2020_paper.pdf
• Code：https://github.com/kingsj0405/ciplab-NTIRE-2020
• Remarks：NTIRE2020极限超分亚军方案CIPLab；Workshops

Nested Scale-Editing for Conditional Image Synthesis

• Paper：http://arxiv.org/abs/2006.02038

MSG-GAN: Multi-Scale Gradients for Generative Adversarial Networks

• Paper：https://arxiv.org/abs/1903.06048v3
• Code：https://github.com/akanimax/msg-stylegan-tf
• Analysis：CVPR2020之MSG-GAN：简单有效的SOTA
• Remarks：NTIRE2020极限超分亚军方案CIPLab；Workshops

视频超分辨率

TDAN: Temporally-Deformable Alignment Network for Video Super-Resolution

• Paper：https://arxiv.org/abs/1812.02898
• Code：https://github.com/YapengTian/TDAN-VSR-CVPR-2020
• Video：https://www.youtube.com/watch?v=eZExENE50I0

Zooming Slow-Mo: Fast and Accurate One-Stage Space-Time Video Super-Resolution

• Paper：https://arxiv.org/abs/2002.11616
• Code：https://github.com/Mukosame/Zooming-Slow-Mo-CVPR-2020
• Analysis：慢镜头变焦：视频超分辨率：CVPR2020Paper解析

Video Super-Resolution With Temporal Group Attention

• Paper：http://openaccess.thecvf.com/content_CVPR_2020/papers/Isobe_Video_Super-Resolution_With_Temporal_Group_Attention_CVPR_2020_paper.pdf

Space-Time-Aware Multi-Resolution Video Enhancement

• 主页：https://alterzero.github.io/projects/STAR.html
• Paper：http://arxiv.org/abs/2003.13170
• Code：https://github.com/alterzero/STARnet

人脸超分辨率

Learning to Have an Ear for Face Super-Resolution

• Paper：https://arxiv.org/abs/1909.12780
• Code：https://github.com/gmeishvili/ear_for_face_super_resolution

Deep Face Super-Resolution With Iterative Collaboration Between Attentive Recovery and Landmark Estimation

• Paper：https://arxiv.org/abs/1812.02898
• Code：https://github.com/YapengTian/TDAN-VSR-CVPR-2020

深度图超分辨率

Channel Attention Based Iterative Residual Learning for Depth Map Super-Resolution

• Paper：https://arxiv.org/abs/2006.01469

光场图像超分辨率

Light Field Spatial Super-Resolution via Deep Combinatorial Geometry Embedding and Structural Consistency Regularization

• Paper：https://arxiv.org/abs/2004.02215
• Code：https://github.com/jingjin25/LFSSR-ATO

高光谱图像超分辨率

Unsupervised Adaptation Learning for Hyperspectral Imagery Super-Resolution

• Paper：http://openaccess.thecvf.com/content_CVPR_2020/papers/Zhang_Unsupervised_Adaptation_Learning_for_Hyperspectral_Imagery_Super-Resolution_CVPR_2020_paper.pdf
• Code：https://github.com/JiangtaoNie/UAL

零样本超分辨率

Meta-Transfer Learning for Zero-Shot Super-Resolution

• Paper：https://arxiv.org/abs/2002.12213
• Code：https://github.com/JWSoh/MZSR

用于超分辨率的数据增广

Rethinking Data Augmentation for Image Super-resolution: A Comprehensive Analysis and a New Strategy

• Paper：http://openaccess.thecvf.com/content_CVPR_2020/html/Yoo_Rethinking_Data_Augmentation_for_Image_Super-resolution_A_Comprehensive_Analysis_and_CVPR_2020_paper.html
• Code：https://github.com/clovaai/cutblur

超分辨率用于语义分割

Dual Super-Resolution Learning for Semantic Segmentation

• Paper：http://openaccess.thecvf.com/content_CVPR_2020/html/Wang_Dual_Super-Resolution_Learning_for_Semantic_Segmentation_CVPR_2020_paper.html
• Code：https://github.com/wanglixilinx/DSRL

其他超分

Explorable Super Resolution

• Paper：https://openaccess.thecvf.com/content_CVPR_2020/papers/Bahat_Explorable_Super_Resolution_CVPR_2020_paper.pdf

2.图像去雨（Image Deraining）

Deep Adversarial Decomposition: A Unified Framework for Separating Superimposed Images

• Paper：http://openaccess.thecvf.com/content_CVPR_2020/html/Zou_Deep_Adversarial_Decomposition_A_Unified_Framework_for_Separating_Superimposed_Images_CVPR_2020_paper.html
• Code：https://github.com/jiupinjia/Deep-adversarial-decomposition
• Demo:http://www-personal.umich.edu/~zzhengxi/zzx_gallery/5946-1min.mp4

Multi-Scale Progressive Fusion Network for Single Image Deraining

• Paper：https://arxiv.org/abs/2003.10985
• Code：https://github.com/kuihua/MSPFN

Syn2Real Transfer Learning for Image Deraining Using Gaussian Processes

• Paper：https://openaccess.thecvf.com/content_CVPR_2020/papers/Yasarla_Syn2Real_Transfer_Learning_for_Image_Deraining_Using_Gaussian_Processes_CVPR_2020_paper.pdf
• Code:https://github.com/rajeevyasarla/Syn2Real

Detail-recovery Image Deraining via Context Aggregation Networks

• Paper：https://openaccess.thecvf.com/content_CVPR_2020/papers/Deng_Detail-recovery_Image_Deraining_via_Context_Aggregation_Networks_CVPR_2020_paper.pdf

3.图像去雾（Image Dehazing）

Domain Adaptation for Image Dehazing

• Paper：https://arxiv.org/abs/2005.04668
• Code：https://github.com/shawnchen63/DA_dahazing

Multi-Scale Boosted Dehazing Network with Dense Feature Fusion

• Paper：https://arxiv.org/abs/2004.13388
• Code：https://github.com/BookerDeWitt/MSBDN-DFF

BidNet: Binocular Image Dehazing Without Explicit Disparity Estimation

• Paper：https://openaccess.thecvf.com/content_CVPR_2020/papers/Pang_BidNet_Binocular_Image_Dehazing_Without_Explicit_Disparity_Estimation_CVPR_2020_paper.pdf

Distilling Image Dehazing With Heterogeneous Task Imitation

• Paper：https://openaccess.thecvf.com/content_CVPR_2020/papers/Hong_Distilling_Image_Dehazing_With_Heterogeneous_Task_Imitation_CVPR_2020_paper.pdf
• Code：https://github.com/FadeoN/Distilling-Image-Dehazing-With-Heterogeneous-Task-Imitation

4.去模糊（Deblurring）

视频去模糊

Cascaded Deep Video Deblurring Using Temporal Sharpness Prior

• Paper：https://arxiv.org/abs/2004.02501
• Code：https://github.com/csbhr/CDVD-TSP
• Homepage：https://csbhr.github.io/projects/cdvd-tsp/index.html

Learning Event-Based Motion Deblurring

• Paper：https://openaccess.thecvf.com/content_CVPR_2020/papers/Jiang_Learning_Event-Based_Motion_Deblurring_CVPR_2020_paper.pdf

Variational-EM-Based Deep Learning for Noise-Blind Image Deblurring

• Paper：https://openaccess.thecvf.com/content_CVPR_2020/papers/Nan_Variational-EM-Based_Deep_Learning_for_Noise-Blind_Image_Deblurring_CVPR_2020_paper.pdf

Efficient Dynamic Scene Deblurring Using Spatially Variant Deconvolution Network With Optical Flow Guided Training

• Paper：https://openaccess.thecvf.com/content_CVPR_2020/papers/Yuan_Efficient_Dynamic_Scene_Deblurring_Using_Spatially_Variant_Deconvolution_Network_With_CVPR_2020_paper.pdf

Deblurring by Realistic Blurring

• Paper：https://openaccess.thecvf.com/content_CVPR_2020/papers/Zhang_Deblurring_by_Realistic_Blurring_CVPR_2020_paper.pdf

Spatially-Attentive Patch-Hierarchical Network for Adaptive Motion Deblurring

• Paper：https://openaccess.thecvf.com/content_CVPR_2020/papers/Suin_Spatially-Attentive_Patch-Hierarchical_Network_for_Adaptive_Motion_Deblurring_CVPR_2020_paper.pdf

Deblurring Using Analysis-Synthesis Networks Pair

• Paper：https://openaccess.thecvf.com/content_CVPR_2020/papers/Kaufman_Deblurring_Using_Analysis-Synthesis_Networks_Pair_CVPR_2020_paper.pdf

5.去噪（Denoising）

A Physics-based Noise Formation Model for Extreme Low-light Raw Denoising

• Paper：https://arxiv.org/abs/2003.12751
• Code：https://github.com/Vandermode/NoiseModel

CycleISP: Real Image Restoration via Improved Data Synthesis

• Paper：https://arxiv.org/abs/2003.07761
• Code：https://github.com/swz30/CycleISP

Self2Self With Dropout: Learning Self-Supervised Denoising From Single Image

• Paper：https://openaccess.thecvf.com/content_CVPR_2020/html/Quan_Self2Self_With_Dropout_Learning_Self-Supervised_Denoising_From_Single_Image_CVPR_2020_paper.html
• Code：https://github.com/scut-mingqinchen/self2self

6.图像恢复（Image Restoration）

Learning Invariant Representation for Unsupervised Image Restoration

• Paper：https://arxiv.org/pdf/2003.12769.pdf
• Code：https://github.com/Wenchao-Du/LIR-for-Unsupervised-IR

Attentive Normalization for Conditional Image Generation

• Paper：https://arxiv.org/abs/2004.03828
• Code：https://github.com/shepnerd/AttenNorm

Bringing Old Photos Back to Life

• Paper：https://arxiv.org/abs/2004.09484
• Code：https://github.com/microsoft/Bringing-Old-Photos-Back-to-Life
• Homepage：http://raywzy.com/Old_Photo/

CycleISP: Real Image Restoration via Improved Data Synthesis

• Paper：https://openaccess.thecvf.com/content_CVPR_2020/papers/Zamir_CycleISP_Real_Image_Restoration_via_Improved_Data_Synthesis_CVPR_2020_paper.pdf
• Code：https://github.com/swz30/CycleISP

Enhanced Blind Face Restoration With Multi-Exemplar Images and Adaptive Spatial Feature Fusion

• Paper：https://openaccess.thecvf.com/content_CVPR_2020/papers/Li_Enhanced_Blind_Face_Restoration_With_Multi-Exemplar_Images_and_Adaptive_Spatial_CVPR_2020_paper.pdf
• Code：https://github.com/csxmli2016/ASFFNet

Disparity-Aware Domain Adaptation in Stereo Image Restoration

• Paper：https://openaccess.thecvf.com/content_CVPR_2020/papers/Yan_Disparity-Aware_Domain_Adaptation_in_Stereo_Image_Restoration_CVPR_2020_paper.pdf

7.图像增强（Image Enhancement）

DeepLPF: Deep Local Parametric Filters for Image Enhancement

• Paper：https://arxiv.org/abs/2003.13985
• Code：https://github.com/sjmoran/deep_local_parametric_filters

Learning for Video Compression With Hierarchical Quality and Recurrent Enhancement

• Paper：https://openaccess.thecvf.com/content_CVPR_2020/papers/Yang_Learning_for_Video_Compression_With_Hierarchical_Quality_and_Recurrent_Enhancement_CVPR_2020_paper.pdf
• Code：https://github.com/RenYang-home/HLVC

Zero-Reference Deep Curve Estimation for Low-Light Image Enhancement

• Paper：https://openaccess.thecvf.com/content_CVPR_2020/papers/Guo_Zero-Reference_Deep_Curve_Estimation_for_Low-Light_Image_Enhancement_CVPR_2020_paper.pdf
• Code：https://github.com/Li-Chongyi/Zero-DCE

From Fidelity to Perceptual Quality: A Semi-Supervised Approach for Low-Light Image Enhancement

• Paper：https://openaccess.thecvf.com/content_CVPR_2020/papers/Yang_From_Fidelity_to_Perceptual_Quality_A_Semi-Supervised_Approach_for_Low-Light_CVPR_2020_paper.pdf
• Code：https://github.com/flyywh/CVPR-2020-Semi-Low-Light

Space-Time-Aware Multi-Resolution Video Enhancement

• Paper：https://openaccess.thecvf.com/content_CVPR_2020/papers/Haris_Space-Time-Aware_Multi-Resolution_Video_Enhancement_CVPR_2020_paper.pdf
• Code：https://github.com/alterzero/STARnet

Learning to Restore Low-Light Images via Decomposition-and-Enhancement

• Paper：https://openaccess.thecvf.com/content_CVPR_2020/papers/Xu_Learning_to_Restore_Low-Light_Images_via_Decomposition-and-Enhancement_CVPR_2020_paper.pdf

8.图像去摩尔纹（Image Demoireing）

Image Demoireing with Learnable Bandpass Filters

• Paper：https://openaccess.thecvf.com/content_CVPR_2020/papers/Zheng_Image_Demoireing_with_Learnable_Bandpass_Filters_CVPR_2020_paper.pdf
• Code：https://github.com/zhenngbolun/Learnbale_Bandpass_Filter

9.图像修复（Inpainting）

Contextual Residual Aggregation for Ultra High-Resolution Image Inpainting

• Paper：https://arxiv.org/abs/2005.09704
• Code：https://github.com/phillips96/CRA-Inpainting

UCTGAN: Diverse Image Inpainting based on Unsupervised Cross-Space

• Paper：http://openaccess.thecvf.com/content_CVPR_2020/papers/Zhao_UCTGAN_Diverse_Image_Inpainting_Based_on_Unsupervised_Cross-Space_Translation_CVPR_2020_paper.pdf

Recurrent Feature Reasoning for Image Inpainting

• Paper：https://openaccess.thecvf.com/content_CVPR_2020/papers/Li_Recurrent_Feature_Reasoning_for_Image_Inpainting_CVPR_2020_paper.pdf
• Code：https://github.com/jingyuanli001/RFR-Inpainting

Prior Guided GAN Based Semantic Inpainting

• Paper：https://openaccess.thecvf.com/content_CVPR_2020/papers/Lahiri_Prior_Guided_GAN_Based_Semantic_Inpainting_CVPR_2020_paper.pdf

3D Photography Using Context-Aware Layered Depth Inpainting

• Paper：https://openaccess.thecvf.com/content_CVPR_2020/papers/Shih_3D_Photography_Using_Context-Aware_Layered_Depth_Inpainting_CVPR_2020_paper.pdf
• Code：https://github.com/vt-vl-lab/3d-photo-inpainting

10.图像质量评价（Image Quality Assessment）

MetaIQA: Deep Meta-Learning for No-Reference Image Quality Assessment

• Paper：https://openaccess.thecvf.com/content_CVPR_2020/papers/Zhu_MetaIQA_Deep_Meta-Learning_for_No-Reference_Image_Quality_Assessment_CVPR_2020_paper.pdf
• Code：https://github.com/zhuhancheng/MetaIQA

Uncertainty-Aware Score Distribution Learning for Action Quality Assessment

• Paper：https://openaccess.thecvf.com/content_CVPR_2020/papers/Tang_Uncertainty-Aware_Score_Distribution_Learning_for_Action_Quality_Assessment_CVPR_2020_paper.pdf
• Code：https://github.com/nzl-thu/MUSDL

Perceptual Quality Assessment of Smartphone Photography

• Paper：https://openaccess.thecvf.com/content_CVPR_2020/papers/Fang_Perceptual_Quality_Assessment_of_Smartphone_Photography_CVPR_2020_paper.pdf
• Code：https://github.com/h4nwei/SPAQ

Adaptive Fractional Dilated Convolution Network for Image Aesthetics Assessment

• Paper：https://openaccess.thecvf.com/content_CVPR_2020/papers/Chen_Adaptive_Fractional_Dilated_Convolution_Network_for_Image_Aesthetics_Assessment_CVPR_2020_paper.pdf

Deep Metric Learning via Adaptive Learnable Assessment

• Paper：https://openaccess.thecvf.com/content_CVPR_2020/papers/Zheng_Deep_Metric_Learning_via_Adaptive_Learnable_Assessment_CVPR_2020_paper.pdf

11.其他多任务

Image Processing Using Multi-Code GAN Prior

• Paper：https://openaccess.thecvf.com/content_CVPR_2020/html/Gu_Image_Processing_Using_Multi-Code_GAN_Prior_CVPR_2020_paper.html
• Code：https://github.com/genforce/mganprior

EventSR: From Asynchronous Events to Image Reconstruction, Restoration, and Super-Resolution via End-to-End Adversarial Learning

• Paper：http://openaccess.thecvf.com/content_CVPR_2020/papers/Wang_EventSR_From_Asynchronous_Events_to_Image_Reconstruction_Restoration_and_Super-Resolution_CVPR_2020_paper.pdf
• Video ：https://www.youtube.com/watch?v=OShS_MwHecs
• Dataset： https://github.com/wl082013/ESIM_dataset

待续更新~

参考

[1] 杜克大学提出 AI 算法，拯救渣画质马赛克秒变高清
[2] CVPR 2020 论文大盘点-超分辨率篇
[3] CVPR2020丨SPSR：基于梯度指导的结构保留超分辨率方法
[4] CVPR2020：USRNet
[5] UDVD：适用于可变降质类型的通用图像超分，附参考代码
[6] NTIRE2020冠军方案RFB-ESRGAN：带感受野模块的超分网络
[7] 超越RCAN，图像超分又一峰：RFANet
[8] #每日五分钟一读#Image Super-Resolution
[9] CVPR 2020 | 几篇GAN在low-level vision中的应用论文
[10] 超100篇！CVPR 2020最全GAN论文梳理汇总！
[11] CVPR2020之MSG-GAN：简单有效的SOTA
[12] CVPR2020-Code
[13] 慢镜头变焦：视频超分辨率：CVPR2020论文解析

相关Low-Level-Vision整理

• Awesome-ECCV2020-Low-Level-Vision