Image Model Blocks
Squeeze-and-Excitation Block
Introduced by Hu et al. in Squeeze-and-Excitation NetworksThe Squeeze-and-Excitation Block is an architectural unit designed to improve the representational power of a network by enabling it to perform dynamic channel-wise feature recalibration. The process is:
- The block has a convolutional block as an input.
- Each channel is "squeezed" into a single numeric value using average pooling.
- A dense layer followed by a ReLU adds non-linearity and output channel complexity is reduced by a ratio.
- Another dense layer followed by a sigmoid gives each channel a smooth gating function.
- Finally, we weight each feature map of the convolutional block based on the side network; the "excitation".
Papers
| Paper | Code | Results | Date | Stars |
|---|
Tasks
| Task | Papers | Share |
|---|---|---|
| Image Classification | 104 | 13.72% |
| Object Detection | 48 | 6.33% |
| Semantic Segmentation | 38 | 5.01% |
| Classification | 37 | 4.88% |
| General Classification | 29 | 3.83% |
| Instance Segmentation | 17 | 2.24% |
| Decoder | 16 | 2.11% |
| Quantization | 13 | 1.72% |
| Multi-Task Learning | 10 | 1.32% |
Usage Over Time
Components
| Component | Type |
|
|---|---|---|
Average Pooling
|
Pooling Operations | |
Convolution
|
Convolutions | |
Dense Connections
|
Feedforward Networks | |
ReLU
|
Activation Functions | |
Sigmoid Activation
|
Activation Functions |
