search

soapisnotfat / super-resolution

collection of super-resolution models & algorithms
Apache License 2.0
744 stars 188 forks source link

problem with super_resolve.py #8

Closed micklexqg closed 5 years ago

micklexqg commented 5 years ago

Hi, thanks for your work. I tried it with 'srcnn' and get a trained model. but when i am a little confused with the result when applying the model with super_resolve.py . input (default in super_resolve.py): image output: image all parameters are set default, such as batchsize=1. why is it so?

soapisnotfat commented 5 years ago

I don't know the exact reason. However, I guess the possible cause is that I convert the image to YCbCr format and only feed Y channel as the input to the network. I'm working on changing the pre-processing so that the input will be RGB instead.

soapisnotfat commented 5 years ago

It seems like RGB configuration cannot preserve the color.

soapisnotfat commented 5 years ago

problem fixed

NagasaiBharat commented 5 years ago

@icpm @micklexqg I am also getting the same problem ,i.e not getting the output in RGB. How can I fix it?

illust commented 5 years ago

@icpm @micklexqg I am also getting the same problem ,i.e not getting the output in RGB. How can I fix it?

Hello, have you solved this problem?