Giters

leftthomas / ESPCN

A PyTorch implementation of ESPCN based on CVPR 2016 paper "Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Network"

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

ESPCN's architecture

kligvasser opened this issue · comments

commented

The suggested model is a bit different than the on in the paper. You got 4 convolutional layers, whereas, there are only 3 in the paper. Have you tried less? Why do you have sigmoid in the output layer?
Thank you very much!
Idan

commented

@kligvasser , you could change that model, it's easy, I use sigmoid to make sure the output image value is between [0,1], you also could use tanh to the last layer, and add 1 then div it by 2 to make sure it between [0,1].