[图像处理任务数据集]

Image Enhancement

• MIT-Adobe FiveK Dataset 单反相机拍摄的raw图像 + 5个人工retouch的tiff图, 共5k张
  
• DPED 手机相机和单反相机增强数据集, 使用sift做对齐, 包括来自三种手机和Canon单反的超过15k张图像对
  

Low Light

• SID
  

• ExDark
  

• CID
  

• VV
  

• LOL
  

• LIME
  

• RENOIR

• ExDARK
  

• SICE
  

• The Extended Yale Face Database B
  

Super Resolution

• DIV2k
  

Real Image Denoising

• DnD
  
• SIDD
  
• PolyU
  
• Renoir
  
• CC
  
• SID
  
• kodak_color
  
• NoiseClinicImages
  

Real Video Denoising

• CRVD dataset Raw视频去噪数据集. 11个室内场景, 5个ISO.
  

Deblurring

• HIDE image motion deblur
  
• GOPRO image motion deblur, 240fps, 1280x720
  
• Deep Video Deblurring(Adobe240fps) video deblur
  
• REDS 视频去模糊, 视频超分, 120fps, 1280x720
  
• BSD real-world video deblur
  

Face Redrawing

• CelebA 人脸属性数据集, 超过200k张图像, 5个关键点, 40个属性
  
• CelebAMask-HQ 3w张人脸解析数据集, 包括19个facial components
  
• Flickr-Faces-HQ Dataset (FFHQ)
  
• PPR10K
  

Reflection Removal

• SIR
  

AWB

• GehlerShi
  
• CUBE+
  

Video datasets

• Vimeo-90k 超分, 插帧, 30fps
  
• Middlebury 光流估计, 视频插帧
  
• UCF101 视频动作识别
  
• X4K1000FPS 4k分辨率,1000fps
  
• DAVIS 视频分割, 1080p, 24fps
  

Conference paper collection

• [CVPR 2022] [CVPR 2022] [CVPR 2021] [CVPR 2020]
• [ECCV 2022]
• [ICCV 2021]

Low level vision conference paper collection

• [CVPR 2022] [CVPR 2021/2020]
• [ECCV 2022] [ECCV 2020]
• [ICCV 2021]

Paper collection on low-level tasks

• [超分论文整理1] [图像/视频超分论文整理2] [图像/视频超分方法TF复现]

• [VideoSuperResolution] A collection of state-of-the-art video or single-image super-resolution architectures, reimplemented in TF

• Low light enhancement [Project 1] [Project 2] [Project 3]

• [image-to-image translation]

• [Deblurring]

• [SimDeblur] 多个deep-learning based图像和视频去模糊Pytorch代码实现

• [Dehazing]

• [Denoising] [Denoising with codes]

[传统图像质量增强的系列blog-1]

[传统图像质量增强的系列blog-2]

[ISP介绍blog]

无法抗拒的图像技术

All in Camera

计算摄影学

AIWalker

GiantPandaCV

极市平台

CVer

[Mobile AI benchmark]

[Waifu2x-Extension-GUI] Image & GIF & Video Super-Resolution and Video Frame Interpolation using DNNs

[Photons to Photos]