Convolutions
Dilated Convolution
Introduced by Yu et al. in Multi-Scale Context Aggregation by Dilated ConvolutionsDilated Convolutions are a type of convolution that “inflate” the kernel by inserting holes between the kernel elements. An additional parameter $l$ (dilation rate) indicates how much the kernel is widened. There are usually $l-1$ spaces inserted between kernel elements.
Note that concept has existed in past literature under different names, for instance the algorithme a trous, an algorithm for wavelet decomposition (Holschneider et al., 1987; Shensa, 1992).
Source: Multi-Scale Context Aggregation by Dilated Convolutions
Papers
| Paper | Code | Results | Date | Stars |
|---|
Tasks
| Task | Papers | Share |
|---|---|---|
| Semantic Segmentation | 163 | 16.48% |
| Reinforcement Learning (RL) | 72 | 7.28% |
| Object Detection | 43 | 4.35% |
| Image Segmentation | 43 | 4.35% |
| Continuous Control | 22 | 2.22% |
| Autonomous Driving | 22 | 2.22% |
| General Classification | 21 | 2.12% |
| Image Classification | 20 | 2.02% |
| Instance Segmentation | 20 | 2.02% |
Usage Over Time
Components
| Component | Type |
|
|---|---|---|
| 🤖 No Components Found | You can add them if they exist; e.g. Mask R-CNN uses RoIAlign |
