SMSR
SMSR copied to clipboard
difference with paper
https://github.com/LongguangWang/SMSR/blob/af9b04ceb5e6820f5df5e372f58f66bb8f85d024/model/smsr.py#L253
I suggest that the subscript of ch_mask may be [:,:,i-1:i,:1]
Supplement: Fig. 5 in the paper shows that the feature should multiply with the channel mask of the last layer before convolutions. But in the code, the feature multiply with the channel mask of the current layer. pls check the code.
Hi @nicetomeetu21, thanks for reporting this issue and we have fixed it.