Table 12 Unseen CUB_200_2011 classification results
From: Multimodal prototype fusion network for paper-cut image classification
Methods | Type | Training | n-shot accuracy (%) | |
|---|---|---|---|---|
5-shot | 16-shot | |||
COOP | Prompt learning | Yes | 33.48 | 35.54 |
CoCoOp | Prompt learning | Yes | 52.10 | 52.30 |
CLIP Zero-shot | Contrastive learning | No | 49.26 | 49.26 |
CLIP-Linear | Deep learning | Yes | 45.45 | 50.66 |
Prototypical Network | Metric learning | No | 62.96 | 68.63 |
Matching Network | Metric learning | No | 51.52 | 62.51 |
AMPFN | Metric learning | Yes | 63.36 | 69.32 |
IMPFN | Metric learning | No | 64.66 | 68.40 |
