[Question] What do these concatenation intend?

[bamps53]

Hi, thanks for the interesting and useful works!

Can you help me to understand this part of the model?
Why do you concat avg pooled one (x2) and kind of normalized one (x3)?
Thanks in advance.

CenterNet-HarDNet/src/lib/models/networks/hardnet.py

Lines 514 to 518 in 6346bd2

	x = self.last_proj(x)
	x = self.last_pool(x)
	x2 = self.avg9x9(x)
	x3 = x/(x.sum((2,3),keepdim=True) + 0.1)
	x = torch.cat([x,x2,x3],1)