search

hitzhangyu / Self-supervised-Image-Enhancement-Network-Training-With-Low-Light-Images-Only

Self-supervised Image Enhancement Network: Training With Low Light Images Only
137 stars 18 forks source link

Max channel for low contrast train images computed using ground-truth? #13

Closed balassbals closed 3 years ago

balassbals commented 3 years ago

Line 71 in main.py (copied below) computes max channel for low contrast images using the ground-truth images (high_im)!! But it is supposed to be self-supervision? Am I missing something?

train_low_data_max_chan = np.max(high_im,axis=2,keepdims=True)

balassbals commented 3 years ago

Line 71 in main.py (copied below) computes max channel for low contrast images using the ground-truth images (_highim)!! But it is supposed to be self-supervision? Am I missing something?

train_low_data_max_chan = np.max(high_im,axis=2,keepdims=True)

Got it. high_im are same as low_im in your code.