Viet-Anh on Software Logo

What is: ResNet-D?

SourceBag of Tricks for Image Classification with Convolutional Neural Networks
Year2000
Data SourceCC BY-SA - https://paperswithcode.com

ResNet-D is a modification on the ResNet architecture that utilises an average pooling tweak for downsampling. The motivation is that in the unmodified ResNet, the 1 × 1 convolution for the downsampling block ignores 3/4 of input feature maps, so this is modified so no information will be ignored