problems about the size of the training patch and receptive field

cszn / IRCNN

Learning Deep CNN Denoiser Prior for Image Restoration (CVPR, 2017) (Matlab)

589 stars 173 forks source link

Open liang233 opened 5 years ago

liang233 commented 5 years ago

Hello, in paper,it said if the size of the training patch is smaller than the receptive field, the performance would decrease. why? thanks