| ID | Backbone | Model | Embedding Size | Modality | Accuracy |
|---|---|---|---|---|---|
| 0 | Conv4 | ProtoNet | 800 | Image | 46.99 |
| - | Conv4 | ProtoNet | 128 | Image | 51.09 +- 0.77 |
| 1 | SentenceEncoder | ProtoNet | 800 | Text | 76.15 |
| - | SentenceEncoder | ProtoNet | 128 | Text | 75.70 |
| 2 | Conv4 | MAML | 800 | Image | 49.75 |
| - | Conv4 | MAML | 128 | Image | 53.16 +- 0.72 |
| 3 | ResNet12 | ProtoNet | 800 | Image | 53.65 |
| - | ResNet12 | ProtoNet | 128 | Image | 58.38 |
|4|Conv4|ProtoNet|128|Image|TODO|
| ID | Backbone | Model | Embedding Size | Modality | Fusion Method | Accuracy |
|---|---|---|---|---|---|---|
| 0 | Conv4 | ProtoNet | 800 | Image | - | 46.99 |
| 5 | Conv4 | ProtoNet | 128 | Image + Text | Mean | TODO |
| 6 | Conv4 | ProtoNet | 128 | Image + Text | FC | TODO |
| 7 | Conv4 | ProtoNet | 128 | Image + Text | Attention (image guided) | TODO |
| 8 | Conv4 | ProtoNet | 128 | Image + Text | Attention (text residual) | TODO |
| 3 | ResNet12 | ProtoNet | 800 | Image | - | 53.65 |
| 9 | ResNet12 | ProtoNet | 128 | Image + Text | Mean | TODO |
| 10 | ResNet12 | ProtoNet | 128 | Image + Text | FC | TODO |
| 11 | ResNet12 | ProtoNet | 128 | Image + Text | Attention (image guided) | TODO |
| 12 | ResNet12 | ProtoNet | 128 | Image + Text | Attention (text residual) | TODO |
| 2 | Conv4 | MAML | 800 | Image | - | 49.75 |
| 13 | Conv4 | MAML | 128 | Image + Text | Mean | TODO |
| 14 | Conv4 | MAML | 128 | Image + Text | FC | TODO |
| 15 | Conv4 | MAML | 128 | Image + Text | Attention (image guided) | TODO |
| 16 | Conv4 | MAML | 128 | Image + Text | Attention (text residual) | TODO |