Image Models
Self-Attention Network
Introduced by Zhao et al. in Exploring Self-attention for Image RecognitionSelf-Attention Network (SANet) proposes two variations of self-attention used for image recognition: 1) pairwise self-attention which generalizes standard dot-product attention and is fundamentally a set operator, and 2) patchwise self-attention which is strictly more powerful than convolution.
Source: Exploring Self-attention for Image Recognition
Papers
| Paper | Code | Results | Date | Stars |
|---|
Tasks
| Task | Papers | Share |
|---|---|---|
| Semantic Segmentation | 2 | 10.00% |
| Super-Resolution | 2 | 10.00% |
| Style Transfer | 2 | 10.00% |
| Crowd Counting | 1 | 5.00% |
| Scene Recognition | 1 | 5.00% |
| Decoder | 1 | 5.00% |
| Real-Time Semantic Segmentation | 1 | 5.00% |
| Self-Driving Cars | 1 | 5.00% |
| Image Dehazing | 1 | 5.00% |
Usage Over Time
Components
| Component | Type |
|
|---|---|---|
| 🤖 No Components Found | You can add them if they exist; e.g. Mask R-CNN uses RoIAlign |
