Regularization
DropBlock
Introduced by Ghiasi et al. in DropBlock: A regularization method for convolutional networksDropBlock is a structured form of dropout directed at regularizing convolutional networks. In DropBlock, units in a contiguous region of a feature map are dropped together. As DropBlock discards features in a correlated area, the networks must look elsewhere for evidence to fit the data.
Source: DropBlock: A regularization method for convolutional networks
Papers
| Paper | Code | Results | Date | Stars |
|---|
Tasks
| Task | Papers | Share |
|---|---|---|
| Object Detection | 51 | 26.29% |
| Real-Time Object Detection | 10 | 5.15% |
| Image Classification | 9 | 4.64% |
| Person Re-Identification | 8 | 4.12% |
| Semantic Segmentation | 6 | 3.09% |
| Autonomous Driving | 5 | 2.58% |
| General Classification | 5 | 2.58% |
| Instance Segmentation | 4 | 2.06% |
| Domain Adaptation | 3 | 1.55% |
Usage Over Time
Components
| Component | Type |
|
|---|---|---|
| 🤖 No Components Found | You can add them if they exist; e.g. Mask R-CNN uses RoIAlign |
