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
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
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
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
深度图超分辨率
Channel Attention Based Iterative Residual Learning for Depth Map Super-Resolution
光场图像超分辨率
Light Field Spatial Super-Resolution via Deep Combinatorial Geometry Embedding and Structural Consistency Regularization
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
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
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
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
3.图像去雾(Image Dehazing)
Domain Adaptation for Image Dehazing
Multi-Scale Boosted Dehazing Network with Dense Feature Fusion
BidNet: Binocular Image Dehazing Without Explicit Disparity Estimation
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
Variational-EM-Based Deep Learning for Noise-Blind Image Deblurring
Efficient Dynamic Scene Deblurring Using Spatially Variant Deconvolution Network With Optical Flow Guided Training
Deblurring by Realistic Blurring
Spatially-Attentive Patch-Hierarchical Network for Adaptive Motion Deblurring
Deblurring Using Analysis-Synthesis Networks Pair
5.去噪(Denoising)
A Physics-based Noise Formation Model for Extreme Low-light Raw Denoising
CycleISP: Real Image Restoration via Improved Data Synthesis
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
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
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
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
UCTGAN: Diverse Image Inpainting based on Unsupervised Cross-Space
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
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
Deep Metric Learning via Adaptive Learnable Assessment
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