Convolutions
Displaced Aggregation Units
Introduced by Tabernik et al. in Spatially-Adaptive Filter Units for Compact and Efficient Deep Neural NetworksDisplaced Aggregation Unit replaces classic convolution layer in ConvNets with learnable positions of units. This introduces explicit structure of hierarchical compositions and results in several benefits:
- fully adjustable and learnable receptive fields through spatially-adjustable filter units
- reduced parameters for spatial coverage efficient inference
- decupling of the parameters from the receptive field sizes
More information can be found here.
Source: Spatially-Adaptive Filter Units for Compact and Efficient Deep Neural Networks
Papers
| Paper | Code | Results | Date | Stars |
|---|
Tasks
| Task | Papers | Share |
|---|---|---|
| Image Classification | 2 | 40.00% |
| Semantic Segmentation | 2 | 40.00% |
| Blind Image Deblurring | 1 | 20.00% |
Usage Over Time
Components
| Component | Type |
|
|---|---|---|
| 🤖 No Components Found | You can add them if they exist; e.g. Mask R-CNN uses RoIAlign |
