Table 14 Classification accuracy of base classes for paper-cutting images under different residual proportion conditions
From: Multimodal prototype fusion network for paper-cut image classification
Initial value | 1-shot(%) | 5-shot(%) | 16-shot(%) |
|---|---|---|---|
0.2 | 82.19 | 89.07 | 90.71 |
0.4 | 76.77 | 87.70 | 90.21 |
0.6 | 72.13 | 87.43 | 90.34 |
0.8 | 67.49 | 86.88 | 87.97 |