Denoising.md

Denoising

• Image Denoising
• Raw Denoising
• Video Denoising
• Signal Processing Based

Image Denoising

• Lighting Every Darkness in Two Pairs: A Calibration-Free Pipeline for RAW Denoising
  Xin Jin, Jia-Wen Xiao, Ling-Hao Han, Chunle Guo, Ruixun Zhang, Xialei Liu, Chongyi Li
  [ICCV 2023] [Project] [Pytorch-Code]
  

• https://github.com/yuanzhi-zhu/DiffPIR
  Yuanzhi Zhu, Kai Zhang, Jingyun Liang, Jiezhang Cao, Bihan Wen, Radu Timofte, Luc Van Gool
  [CVPR 2023] [Project] [Pytorch-Code]
  

• sRGB Real Noise Synthesizing with Neighboring Correlation-Aware Noise Model
  Zixuan Fu, Lanqing Guo, Bihan Wen
  [CVPR 2023] [Pytorch-Code]
  

• Practical Blind Denoising via Swin-Conv-UNet and Data Synthesis
  Kai Zhang, Yawei Li, Jingyun Liang, Jiezhang Cao, Yulun Zhang, Hao Tang, Radu Timofte, Luc Van Gool
  [arXiv 2203] [Pytorch-Code]
  [SCUNet]

• Fast and High-quality Image Denoising via Malleable Convolutions
  Yifan Jiang, Bartlomiej Wronski, Ben Mildenhall, Jonathan Barron, Zhangyang Wang, Tianfan Xue
  [ECCV 2022] [Project]
  [MalleConv] [★★] 将图像/feature下采样四倍, 预测dynamic kernel, 另外网络采用unnet的形式, 总体计算量较小, 且指标较高.

• CVF-SID: Cyclic multi-Variate Function for Self-Supervised Image Denoising by Disentangling Noise from Image
  Reyhaneh Neshatavar, Mohsen Yavartanoo, Sanghyun Son, Kyoung Mu Lee
  [CVPR 2022] [Pytorch-Code]
  

• AP-BSN: Self-Supervised Denoising for Real-World Images via Asymmetric PD and Blind-Spot Network
  Wooseok Lee, Sanghyun Son, Kyoung Mu Lee
  [CVPR 2022] [Pytorch-Code]
  

• Blind2Unblind: Self-Supervised Image Denoising with Visible Blind Spots
  Zejin Wang, Jiazheng Liu, Guoqing Li, Hua Han
  [CVPR 2022] [Pytorch-Code]
  

• IDR: Self-Supervised Image Denoising via Iterative Data Refinement
  Yi Zhang, Dasong Li, Ka Lung Law, Xiaogang Wang, Hongwei Qin, Hongsheng Li
  [CVPR 2022] [Pytorch-Code]
  

• Practical Blind Denoising via Swin-Conv-UNet and Data Synthesis
  Kai Zhang, Yawei Li, Jingyun Liang, Jiezhang Cao, Yulun Zhang, Hao Tang, Radu Timofte, Luc Van Gool
  [arXiv 2203] [Pytorch-Code]
  [SCUNet]

• FBI-Denoiser: Fast Blind Image Denoiser for Poisson-Gaussian Noise
  Jaeseok Byun, Sungmin Cha, Taesup Moon
  [CVPR 2021] [Pytorch-Code]
  

• Invertible Denoising Network: A Light Solution for Real Noise Removal
  Yang Liu, Zhenyue Qin, Saeed Anwar, Pan Ji, Dongwoo Kim, Sabrina Caldwell, Tom Gedeon
  [CVPR 2021] [Pytorch-Code]
  

• Deep Denoising of Flash and No-Flash Pairs for Photography in Low-Light Environments
  Zhihao Xia, Michaël Gharbi, Federico Perazzi, Kalyan Sunkavalli, Ayan Chakrabarti
  [CVPR 2021] [TF-Code]
  [deepfnf] [★] 基于"Basis Prediction Networks for Effective Burst Denoising with Large Kernels"设计网络, 另外有一个scale map, 用于从flash image中得到高频细节, 乘到滤波后的non flash图像中.

• NBNet: Noise Basis Learning for Image Denoising with Subspace Projection
  Shen Cheng, Yuzhi Wang, Haibin Huang, Donghao Liu, Haoqiang Fan, Shuaicheng Liu
  [CVPR 2021] [Code]
  [★] UNet结构, 利用decoder产生的高层特征预测basis及投影矩阵, 用于将低层特征投影已达到降噪目的.

• Neighbor2Neighbor: Self-Supervised Denoising from Single Noisy Images
  Tao Huang, Songjiang Li, Xu Jia, Huchuan Lu, Jianzhuang Liu
  [CVPR 2021] [Pytorch-Code]
  

• Beyond Joint Demosaicking and Denoising
  SMA Sharif, Rizwan Ali Naqvi, Mithun Biswas)
  [CVPRW 2021] [Pytorch-Code]
  [BJDD]

• CycleISP: Real Image Restoration via Improved Data Synthesis
  Syed Waqas Zamir, Aditya Arora, Salman Khan, Munawar Hayat, Fahad Shahbaz Khan, Ming-Hsuan Yang, Ling Shao
  [CVPR 2021 Oral] [Pytorch-Code]
  [★] 提出了一个从sRGB到RAW相互转换的网络, 在Raw图像上注入高斯噪声用于生成RGB噪声样本.

• Self2Self With Dropout: Learning Self-Supervised Denoising From Single Image
  Yuhui Quan, Mingqin Chen, Tongyao Pang, Hui Ji
  [CVPR 2021] [TF-Code]
  

• Transfer Learning from Synthetic to Real-Noise Denoising with Adaptive Instance Normalization
  Yoonsik Kim, Jae Woong Soh, Gu Yong Park, Nam Ik Cho
  [CVPR 2020] [TF-Code]
  [AINDNet]

• Memory-Efficient Hierarchical Neural Architecture Search for Image Denoising
  Haokui Zhang, Ying Li, Hao Chen, Chunhua Shen
  [CVPR 2020]
  

• Unpaired Learning of Deep Image Denoising
  Xiaohe Wu, Ming Liu, Yue Cao, Dongwei Ren, Wangmeng Zuo
  [ECCV 2020] [Pytorch-Code]
  [DBSN]

• Reconstructing the Noise Variance Manifold for Image Denoising
  Ioannis Marras, Grigorios G. Chrysos, Ioannis Alexiou, Gregory Slabaugh, Stefanos Zafeiriou
  [ECCV 2020] [Pytorch-Code]
  

• Burst Denoising via Temporally Shifted Wavelet Transforms
  Xuejian Rong, Denis Demandolx, Kevin Matzen, Priyam Chatterjee, Yingli Tian
  [ECCV 2020] [Pytorch-Code]
  

• Burst Denoising via Temporally Shifted Wavelet Transforms
  Majed El Helou, Ruofan Zhou, Sabine Süsstrunk
  [ECCV 2020] [Pytorch-Code]
  [SFM]

• BLearning Graph-Convolutional Representations for Point Cloud Denoising
  Francesca Pistilli, Giulia Fracastoro, [Diego Valsesia](https://ipl.polito.it/people/valsesia/, Enrico Magli
  [ECCV 2020] [TF-Code]
  [GPDNet]

• Spatial Hierarchy Aware Residual Pyramid Network for Time-of-Flight Depth Denoising
  Guanting Dong, Yueyi Zhang, Zhiwei Xiong
  [ECCV 2020] [TF-Code]
  

• A Decoupled Learning Scheme for Real-world Burst Denoising from Raw Images
  Zhetong Liang, Shi Guo, Hong Gu, Huaqi Zhang, Lei Zhang
  [ECCV 2020]
  

• Robust and On-the-fly Dataset Denoising for Image Classification
  Jiaming Song, Lunjia Hu, Michael Auli, Yann Dauphin, Tengyu Ma
  [ECCV 2020]
  

• Spatial-Adaptive Network for Single Image Denoising
  Meng Chang, Qi Li, Huajun Feng, Zhihai Xu
  [ECCV 2020] [Pytorch-Code]
  [SADNet]

• Noise2Same: Optimizing A Self-Supervised Bound for Image Denoising
  Yaochen Xie, Zhengyang Wang, Shuiwang Ji
  [NIPS 2020] [TF-Code]
  

• Patch2Self: Denoising Diffusion MRI with Self-Supervised Learning
  Shreyas Fadnavis, Joshua Batson, Eleftherios Garyfallidis
  [NIPS 2020] [Code]
  

• Real image denoising with feature attention
  Saeed Anwar, Nick Barnes
  [ICCV 2019 Oral] [Code]
  [RIDNet] [★☆] 1) 提出了一个端到端的去噪网络, 基于channel attention和skip connection. 在真是图像上测试效果不错, 速度一般. 2) 作为一篇Oral来说感觉创新点和理论论述都一般, 也没有解释为什么提出的网络对真是图像去噪效果好. 3) 如果需要, 参考网络流程图和代码即可.
  

• Toward Convolutional Blind Denoising of Real Photographs
  Shi Guo, Zifei Yan, Kai Zhang, Wangmeng Zuo, Lei Zhang, Jonathan T. Barron
  [CVPR 2019] [Code]
  [CBDNet] [★☆] 1) 大致浏览. 采用一个FCN估计噪声level, 噪声level map与输入concat然后输入一类似U-Net的网络去噪. 2) 可以学习其网络和训练细节.
  

• FC-AIDE: Fully Convolutional Adaptive Image Denoiser
  Sungmin cha, Taesup Moon
  [CVPR 2019] [TF-Code]
  

• FOCNet: A Fractional Optimal Control Network for Image Denoising
  Xixi Jia, Sanyang Liu, Xiangchu Feng, Lei Zhang
  [CVPR 2019] [Matlab-Code]
  

• Variational Denoising Network: Toward Blind Noise Modeling and Removal
  Zongsheng Yue, Hongwei Yong, Qian Zhao, Deyu Meng, Lei Zhang
  [NeurIPS 2019] [Pytorch-Code]
  [VDNet]

• High-Quality Self-Supervised Deep Image Denoising
  Samuli Laine, Tero Karras, Jaakko Lehtinen, Timo Aila
  [NeurIPS 2019] [TF-Code]
  

• Learning Deep CNN Denoiser Prior for Image Restoration
  Kai Zhang, Wangmeng Zuo, Shuhang Gu, Lei Zhang
  [CVPR 2017] [Matlab-Code]
  [IRCNN] [★] 大致浏览, 提出了一个结构简单的CNN去噪器, 可以为基于模型的优化方法提供有效的prior, 还可以用于求解其它图像恢复的逆问题

Raw Denoising

• Learnability Enhancement for Low-light Raw Denoising: Where Paired Real Data Meets Noise Modeling
  Hansen Feng, Lizhi Wang, Yuzhi Wang, Hua Huang
  [MM 2022] [Pytorch-Code]
  [PMN]

• Efficient Burst Raw Denoising with Variance Stabilization and Multi-frequency Denoising Network
  Dasong Li, Yi Zhang, Ka Lung Law, Xiaogang Wang, Hongwei Qin, Hongsheng Li
  [IJCV 2022]
  [★★] 1) 提出将输入进行变换, 从Poisson-Gaussian distribution变换到与gain和信号强度无关的固定方差高斯分布; 2) 提出了一个轻量级的上/下采样去噪网络, 序列化地处理N个输入帧, 使网络能在端侧运行; 3) 对齐部分使用block-matching和homography flow, 在4级金字塔上corase-to-fine进行; 4) 从实验结果上看, 提出的变换与直接concat噪声map和k-sigma方法相比, 指标稍高一点, 可视化结果细节保留更多.

• Rethinking Noise Synthesis and Modeling in Raw Denoising
  Yi Zhang, Hongwei Qin, Xiaogang Wang, Hongsheng Li
  [ICCV 2021] [Pytorch-Code]
  [★] 将噪声源分解为信号相关和信号无关两部分, 信号相关部分用一般方法建模为泊松噪声, 信号无关方法从black帧种采样, 并证明采样时采用高bit数据更好.

• A Physics-based Noise Formation Model for Extreme Low-light Raw Denoising
  Kaixuan Wei, Ying Fu, Jiaolong Yang, Hua Huang
  [CVPR 2020 Oral] [Pytorch-Code & Dataset]
  [ELD] [★★] 较为全面的分析了相机噪声来源, 并据此提出了一个高度模拟真实的噪声生成模型.

• Basis Prediction Networks for Effective Burst Denoising with Large Kernels
  Zhihao Xia, Federico Perazzi, Michael Gharbi, Kalyan Sunkavalli, Ayan Chakrabarti
  [CVPR 2020] [Project]
  [BPN]

• Practical Deep Raw Image Denoising on Mobile Devices
  Yuzhi Wang, Haibin Huang, Qin Xu, Jiaming Liu, Yiqun Liu, Jue Wang
  [ECCV 2020 Spotlight] [Pytorch-Code]
  [PMRID] [★★] 分析了噪声来源, 提出用k-sigma变换消除iso对噪声水平的影响, 并提出了标定相机噪声参数的方法.

• Unprocessing Images for Learned Raw Denoising
  Tim Brooks, Ben Mildenhall, Tianfan Xue, Jiawen Chen, Dillon Sharlet, Jonathan T. Barron
  [CVPR 2019] [TF-Code]
  [★☆] 1) 提出了一个通过unprocess ISP流程而生成更真实去噪样本的框架, 可以用任意图像生成真实的训练样本, 以提高模型性能. 2) 对于sRGB图像, 根据ISP流程, 将其逐步逆运算位raw image, 在此基础上加的噪声更符合真实噪声. 3) 推断时, 要先把sRGB转换为raw image, 再经过网络处理, 最后再进行正向的ISP恢复为sRGB.
  

• Burst Denoising with Kernel Prediction Networks
  Ben Mildenhall, Jonathan T. Barron, Jiawen Chen, Dillon Sharlet, Ren Ng, Robert Carroll
  [CVPR 2018 Oral] [Project] [TF-Code] [Unofficial-Pytorch-Code]
  [KPN] [★★] 1) 提出了一个从普通RGB通过逆ISP, 加噪等流程生成raw burst denoise数据对的流程 (KPN数据集); 2) 提出将输入与噪声map(标准差)一起输入网络, 改善了性能; 3) 提出了一个预测kernel的网络, 对每个位置预测若然个去噪核, 并集成得到去噪结果.

Video Denoising

• Dancing under the stars: video denoising in starlight
  Kristina Monakhova, Stephan R. Richter, Laura Waller, Vladlen Koltun
  [CVPR 2022 Oral] [Project] [Pytorch-Code]
  

• Patch Craft: Video Denoising by Deep Modeling and Patch Matching
  Gregory Vaksman, Michael Elad, Peyman Milanfar
  [ICCV 2021] [Pytorch-Code]
  [PaCNet] [★] 大致浏览. 在前后帧提取相似patch作为网络输入, 后面加了一个时域滤波网络保证时间一致性, 可能速度会很慢?

• Efficient Multi-Stage Video Denoising with Recurrent Spatio-Temporal Fusion
  Matteo Maggioni, Yibin Huang, Cheng Li, Shuai Xiao, Zhongqian Fu, Fenglong Song
  [CVPR 2021] [Unofficial-Pytorch-Code]
  [EMVD] [★★] 轻量级视频去噪, 效果与复杂模型效果相当. 首先用线性变换将raw图像在颜色-亮度和频率上分解; 第二步利用前一帧去噪结果与当前帧融合, 初步去噪; 第三步对初步去噪的图像再次进行去噪; 第四步将两次去噪的结果结合进行refine. 融合和refine是通过预测fusion map完成的.

• FastDVDnet: A Very Fast Deep Video Denoising algorithm
  Matias Tassano, Julie Delon, Thomas Veit
  [CVPR 2020] [Pytorch-Code]
  

• Supervised Raw Video Denoising With a Benchmark Dataset on Dynamic Scenes
  Huanjing Yue, Cong Cao, Lei Liao, Ronghe Chu, Jingyu Yang
  [CVPR 2020] [Pytorch-Code]
  [RViDeNet] [★★] 提出了CRVD dataset, 用于Raw去噪.

Signal Processing Based

• Practical Signal-Dependent Noise Parameter Estimation From a Single Noisy Image
  Xinhao Liu, Masayuki Tanaka, Masatoshi Okutomi
  [TIP 2014]
  [★★] 将噪声模型建模为异方差高斯分布, 并通过小波和最小二乘求得噪声参数, 并分析了clipping对噪声建模的影响.

• Practical Poissonian-Gaussian Noise Modeling and Fitting for Single-Image Raw-Data
  Alessandro Foi, Mejdi Trimeche, Vladimir Katkovnik, Karen Egiazarian
  [TIP 2008]
  [★★] 统计方差和均值, 从单张图像种拟合出噪声参数