Cross-View-Gait-Based-Human-Identification-with-Deep-CNNs icon indicating copy to clipboard operation
Cross-View-Gait-Based-Human-Identification-with-Deep-CNNs copied to clipboard

Published 20 hours ago • verified publisher icon fucusy

questions about the model you built

Open michaelHaha opened this issue 6 years ago • 4 comments

Hi, I am also from Beihang University. I am currently a junior student and also try to reproduce this paper.

However, I have some doubts about the model you built. It seems that the model you built is MT model. And the third convolution layer is devised to learn to perform 'subtraction' operation with outputs of two-channel convolution below. 1 However, in your code, it seems that you just subtract the outputs of the two-channel convolution manually and then do convolution on this subtraction. local merge = nn.Sequential(); merge:add(nn.CSubTable()); merge:add(nn.Abs()) I suppose there is something wrong here. What do you think?

michaelHaha avatar May 30 '18 07:05 michaelHaha

What do you mean "learn to perform 'subtraction' operation"?

fucusy avatar May 30 '18 07:05 fucusy

I mean this, 3 I think it means that you cannot subtract the outputs of two columns directly.

michaelHaha avatar May 30 '18 12:05 michaelHaha

I see. you are right. Hopefully, you can add your implementation based mine and create a pull request.

fucusy avatar May 31 '18 09:05 fucusy

Ok, I will.

michaelHaha avatar Jun 02 '18 02:06 michaelHaha