Skip to content

Kdoger/MIRNet2Paddle

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

6 Commits
 
 
 
 

Repository files navigation

Learning Enriched Features for Real Image Restoration and Enhancement

模型描述:MIRNet模型由卷积层(Conv)和多个递归残差组(RRGs)构成,其中卷积层用用于提取输入数据的低层特征,然后提取到的这些特征图经过RRGs产生深度特征。其中每个RRG由多个多尺度残差快(MRB)构成,在MRB中一个卷积流变成多个卷积流,并通过SKFF模块实现多分辨率的信息交流;通过DAU捕获空间和通道维度的上下文信息,最终又聚合成一个卷积流。

参考代码在SIDD数据集上的验证结果:PSNR=39.72  SSIM=0.9194 image

About

Implement MIRNet with Paddle

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages