文献集合 之 Perceptual Quality Methods、Information Utilization Method
文献集合 之 Perceptual Quality Methods、Information Utilization Method
来自综述文章:《A systematic survey of deep learning-based single-image super-resolution》,2024,4月
文章链接:A Systematic Survey of Deep Learning-Based Single-Image Super-Resolution | ACM Computing Surveys
文献集合 之 Efficient Network / Mechanism Design Methods
文献集合 之 Efficient Network / Mechanism Design Methods
来自综述文章:《A systematic survey of deep learning-based single-image super-resolution》,2024,4月
文章链接:A Systematic Survey of Deep Learning-Based Single-Image Super-Resolution | ACM Computing Surveys
超分辨率第十一章-SR&Diffusion_model
超分辨率第十一章-SR&Diffusion_model
Denoising Diffusion Probabilistic Models 于2020年发布。
参考文献:Denoising Diffusion Probabilistic Models
此后在超分领域出现了两篇以扩散模型为基础的文章:
参考文献:
SR3: Image Super-Resolution via Iterative Refinement (2021)
SRDiff: Single Image Super-Resolution with Diffusion Probabilistic Models (2021)
深度学习第六章-VAE架构
深度学习第六章-VAE架构
VAE(变分自编码器)为一个经典的生成式模型,属于编码器-解码器结构。
编码器用于采用复杂部分的特征,之后将复杂分布转换到简单分布的隐空间(如高斯分布)。
解码器采样高斯分布中的数据,生成符合原始分布的图像。
参考视频:
参考博客:
超分辨率第十章-SR&NormalizingFlow
超分辨率第十章-SR&NormalizingFlow
本文介绍标准化流(Normalizing Flow)以及其对应的超分辨率模型SRflow
参考文献: Normalizing Flows: An Introduction and Review of Current Methods
参考博客:SRFlow: Learning the Super-Resolution Space with Normalizing Flow
超分辨率第八章-SR&transformer
超分辨率第八章-SR&transformer
超分辨率中的transformer-应用于MRI
文献一:(CFTN) 3d Cross-Scale Feature Transformer Network for Brain Mr Image Super-Resolution:ICASSP 2022会议
文献二: (ASFT) Adjacent slices feature transformer network for single anisotropic 3D brain MRI image super-resolution:《Biomedical Signal Processing and Control》 期刊,2022
文献三:(CRFM) Cross-Modality Reference and Feature Mutual-Projection for 3D Brain MRI Image Super-Resolution 期刊,2024
超分辨率第七章-SRFBN
超分辨率第七章-SRFBN
SRFBN发表于2019年,引入了反馈网络机制,不会增加额外的参数,并且多次回传相当于加深了网络。
论文地址:Feedback Network for Image Super-Resoluition
MRI论文:A trusted medical image super-resolution method based on feedback adaptive weighted dense network
参考博客:【CVPR2019】超分辨率文章,SRFBN: Feedback Network for Image Super-Resoluition
代码位置:F:\Github下载\SRFBN_CVPR19-master
超分辨率第六章-RCAN
超分辨率第六章-RCAN
RCAN发表于2018年,引入了注意力机制:Channel Attention (CA)
论文地址:Image Super-Resolution Using Very Deep Residual Channel Attention Networks
参考博客
代码位置:F:\Github下载\RCAN-pytorch-master