Pooling Operations
Soft Pooling
Introduced by Stergiou et al. in Refining activation downsampling with SoftPoolSoftPool: a fast and efficient method that sums exponentially weighted activations. Compared to a range of other pooling methods, SoftPool retains more information in the downsampled activation maps. More refined downsampling leads to better classification accuracy.
Source: Refining activation downsampling with SoftPool
Papers
| Paper | Code | Results | Date | Stars |
|---|
Tasks
| Task | Papers | Share |
|---|---|---|
| Federated Learning | 1 | 33.33% |
| Image Forensics | 1 | 33.33% |
| Action Recognition | 1 | 33.33% |
Usage Over Time
Components
| Component | Type |
|
|---|---|---|
| 🤖 No Components Found | You can add them if they exist; e.g. Mask R-CNN uses RoIAlign |
