Viet-Anh on Software Logo

What is: ResNeSt?

SourceResNeSt: Split-Attention Networks
Year2000
Data SourceCC BY-SA - https://paperswithcode.com

A ResNest is a variant on a ResNet, which instead stacks Split-Attention blocks. The cardinal group representations are then concatenated along the channel dimension: V=ConcatV = \text{Concat}{V1,V2,VKV^{1},V^{2},\cdots{V}^{K}}. As in standard residual blocks, the final output YY of otheur Split-Attention block is produced using a shortcut connection: Y=V+XY=V+X, if the input and output feature-map share the same shape. For blocks with a stride, an appropriate transformation T\mathcal{T} is applied to the shortcut connection to align the output shapes: Y=V+T(X)Y=V+\mathcal{T}(X). For example, T\mathcal{T} can be strided convolution or combined convolution-with-pooling.