Dense-Scale-Network-for-Crowd-Counting icon indicating copy to clipboard operation
Dense-Scale-Network-for-Crowd-Counting copied to clipboard

Published 20 hours ago • verified publisher icon rongliangzi

Multi-scale density level consistency loss

Open victorjourne opened this issue 3 years ago • 1 comments

Thanks again for your work! Regarding the Multi-scale density level consistency loss, I think the L1 loss has to be reduced by sum instead of mean (default pytorch paramater). Indeed, even if it it not specified clearly in the paper, as the norm is divided by the square of the average pooling output size, it is more logical to sum every component of the average pooling operation than average them.

What do you think?

victorjourne avatar Oct 28 '21 10:10 victorjourne

Sorry for the late reply. The changes are reasonable, and I have merged to master branch. Thanks for your contribution!

rongliangzi avatar Oct 31 '21 07:10 rongliangzi