**題目
**方向
**鏈結
源**相關資料
across scales & across dimensions: temporal super-resolution using deep internal learning
運動模糊超分辨？
鏈結comming soon
鏈結component divide-and-conquer for real-world image super-resolution
真實世界影象超分辨（分治方法）
cross-attention in coupled unmixing nets for unsupervised hyperspectral super-resolution
利用光譜***（沒見過的方向）
鏈結鏈結
face super-resolution guided by 3d facial priors
人臉超分辨
fast adaptation to super-resolution networks via meta-learning
影象超分辨+其他方法（元學習）
鏈結journey towards tiny perceptual super-resolution
小型感知sr模型，模型小型化，移動端部署
鏈結latticenet: towards lightweight image super-resolution with lattice block
輕量級sr模型
learning with privileged information for efficient image super-resolution
影象超分辨
鏈結鏈結
mucan: multi-correspondence aggregation network for video super-resolution
pams: quantized super-resolution via parameterized max scale
輕量化？（沒找到**，所以不太清楚）
plugnet: degradation aware scene text recognition supervised by a pluggable super-resolution unit
文字超分辨
鏈結scene text image super-resolution in the wild
文字超分辨（戶外場景）
鏈結鏈結
single image super-resolution via a holistic attention network
影象超分辨（網路結構方面的工作）
鏈結spatial-angular interaction for light field image super-resolution
新型光場影象超分辨演算法
鏈結鏈結
srflow: learning the super-resolution space with normalizing flow
影象超分辨（超分空間）
鏈結鏈結
stochastic frequency masking to improve super-resolution and denoising networks
影象超分辨
鏈結texture hallucination for large-factor painting super-resolution
影象超分辨（繪畫場景）
鏈結towards content-independent multi-reference super-resolution: adaptive pattern matching and feature aggregation
影象超分辨（與內容無關的影象超分）
varsr: variational super-resolution network for very low resolution images
影象超分辨（超低解析度場景）
video super-resolution with recurrent structure-detail network
zero-shot image super-resolution with depth guided internal degradation learning
影象超分辨（對zero-shot方法的改進）
invertible image rescaling
影象超分辨（可逆影象縮放）
鏈結鏈結
鏈結**題目
狀態across scales & across dimensions: temporal super-resolution using deep internal learning
component divide-and-conquer for real-world image super-resolution
cross-attention in coupled unmixing nets for unsupervised hyperspectral super-resolution
face super-resolution guided by 3d facial priors
fast adaptation to super-resolution networks via meta-learning
learning with privileged information for efficient image super-resolution
pams: quantized super-resolution via parameterized max scale
plugnet: degradation aware scene text recognition supervised by a pluggable super-resolution unit
scene text image super-resolution in the wild
single image super-resolution via a holistic attention network
spatial-angular interaction for light field image super-resolution
srflow: learning the super-resolution space with normalizing flow
stochastic frequency masking to improve super-resolution and denoising networks
texture hallucination for large-factor painting super-resolution
towards content-independent multi-reference super-resolution: adaptive pattern matching and feature aggregation
varsr: variational super-resolution network for very low resolution images
zero-shot image super-resolution with depth guided internal degradation learning
invertible image rescaling
超解析度參考
數字影象的儲存 影象四種型別 按照顏色 灰度劃分 影象超解析度及相關知識 簡介 超解析度知識筆記 學習筆記之 基於深度學習的影象超解析度重構 帶打包 兩篇乾貨滿滿的csdn博文 從srcnn到edsr，總結深度學習端到端超解析度方法發展歷程 影象超分辨之zssr 基於dl的超解析度 去噪近幾年 一篇...
DRRN超解析度
全名 image super resolution via deep recursive residual network code 這是一篇cvpr2017的關於影象超解析度的文章，相較於vdsr取得了更好的結果，從結構上看，還是之前的基調，更深的網路結構 52層 同時採取了遞迴操作，遞迴模組中存...
超解析度重建總結
1.srcnn 2,3改進 開山之作，三個卷積層，輸入影象是低解析度影象經過雙三次 bicubic 插值和高解析度乙個尺寸後輸入cnn。影象塊的提取和特徵表示，特徵非線性對映和最終的重建。使用均方誤差 mse 作為損失函式。2.fsrcnn 特徵提取 低解析度影象，選取的核9 9設定為5 5。收縮 ...