Skip to content
/ xUnit Public
forked from kligvasser/xUnit

Learning a Spatial Activation Function for Efficient Image Restoration

Notifications You must be signed in to change notification settings

bulbgod/xUnit

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

31 Commits
 
 
 
 
 
 
 
 

Repository files navigation

xUnit

Learning a Spatial Activation Function for Efficient Image Restoration.

Please refer our paper for more details.

Dependencies

  • python (tested with 3.5)
  • PyTorch >= 0.2.0

Code

Clone this repository into any place you want.

git clone https://github.com/kligvasser/xUnit
cd xUnit

Results

Gaussian Denoising

The average PSNR in [dB] attained by several state of the art denoising algorithms on the BSD68:

Methods BM3D WNNM EPLL MLP DnCNN-S xDnCNN
# Parameters - - - - 555K 303K
σ=25 28.56 28.82 28.68 28.95 29.22 29.21
σ=50 25.62 25.87 25.67 26.01 26.23 26.26

Single Image Super Resolution

The average PSNR in [dB] attained in the task of 3× and 4× SR on BSD100 dataset:

Methods SRCNN xSRCNNc xSRCNNf SRResNet xSRResNet
# Parameters 57K 44K 32K 1.546M 1.155M
28.41 28.54 28.53 - -
26.90 27.04 27.06 27.58 27.61

About

Learning a Spatial Activation Function for Efficient Image Restoration

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 93.8%
  • MATLAB 6.2%