search

layumi / 2016_super_resolution

ICCV2015 Image Super-Resolution Using Deep Convolutional Networks
MIT License
86 stars 43 forks source link

I have one more question about Filter Size on each layer. #8

Open star4s opened 7 years ago

star4s commented 7 years ago

I have one more question about Filter Size on each layer. When I read the reference paper, "Image Super-Resolution Using Deep Convolutional Networks" on 7 and 8 page, it mentioned about filter size 9-5-5. This means that first 9 is size of the first layer and second 5 is size of the second layer, and third 5 is size of the third layer. How can I tuning the size of the each layer? I would like to set up the filter size 9-5-5. What is your filter size? thank you for your attention.

layumi commented 7 years ago

You may refer to https://github.com/layumi/2016_super_resolution/blob/master/SRnet.m I use 9-1-5. Note that if you change to 9-5-5, you should change the output size in the training code, since the output size will be 4pixel smaller. https://github.com/layumi/2016_super_resolution/blob/master/train_SRnet.m#L76