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: $V = \text{Concat}${$V^{1},V^{2},\cdots{V}^{K}$}. As in standard residual blocks, the final output $Y$ of otheur Split-Attention block is produced using a shortcut connection: $Y=V+X$, if the input and output feature-map share the same shape. For blocks with a stride, an appropriate transformation $\mathcal{T}$ is applied to the shortcut connection to align the output shapes: $Y=V+\mathcal{T}(X)$. For example, $\mathcal{T}$ can be strided convolution or combined convolution-with-pooling.
Source: ResNeSt: Split-Attention NetworksPaper | Code | Results | Date | Stars |
---|
Task | Papers | Share |
---|---|---|
Instance Segmentation | 3 | 17.65% |
Medical Image Segmentation | 2 | 11.76% |
Image Classification | 2 | 11.76% |
Object Detection | 2 | 11.76% |
Semantic Segmentation | 2 | 11.76% |
Fake News Detection | 1 | 5.88% |
Anomaly Detection | 1 | 5.88% |
Image Segmentation | 1 | 5.88% |
Retinal Vessel Segmentation | 1 | 5.88% |
Component | Type |
|
---|---|---|
1x1 Convolution
|
Convolutions | |
Convolution
|
Convolutions | |
Residual Connection
|
Skip Connections | |
Split Attention
|
Image Model Blocks |