About Local Attention Module(LAM) and the code

[Urian-oy]

Hi Bro,it's me again.
I read your paper again and plan to share your idea, but a little confuse.

I find that LAM don't match the code.The structure of the LAM module in the paper show the low-frequency portion of the image feature is not multiplied by the weight.But I found that the low-frequency part of your code is multiplied by the weight parameter.I show you your code and the LAM picture.

[c-yn]

Hi, the multiplication parameters are implemented in the previous line of the arrow.

The lamb_l/h are parameters to fuse results from two branches.