collections of excellent feature extractor
Download pre-trained weight
| Paper | Link |
|---|---|
| CrossFormer: A Versatile Vision Transformer Hinging on Cross-scale Attention | 1 |
| BiFormer: Vision Transformer with Bi-Level Routing Attention | 2 |
| HorNet: Efficient High-Order Spatial Interactions with Recursive Gated Convolutions | 3 |
| FcaNet: Frequency Channel Attention Networks | 4 |
| DilateFormer: Multi-Scale Dilated Transformer for Visual Recognition | 5 |
- use pytorch 2.X to run Hornet
