TABLE 2.
Classification performance of FSCC and SOTA methods on the synthetic dataset. The classification performance is measured by classification accuracy, followed by the number of labeled training macromolecules in parentheses.
| Methods | SHREC19 | SHREC21 | ||
|---|---|---|---|---|
| 5-Way-5-Shot | 5-Way-1-Shot | 5-Way-5-Shot | 5-Way-1-Shot | |
| Baseline (Chen et al., 2019) | 73.14 ± 0.47%(50) | 65.12 ± 0.74%(10) | 71.36 ± 1.19%(50) | 59.15 ± 1.70%(10) |
| Baseline++ (Chen et al., 2019) | 75.32 ± 0.37%(50) | 66.65 ± 0.67%(10) | 73.64 ± 1.04%(50) | 65.19 ± 1.02%(10) |
| FSCC (ours) | 77.03 ± 1.21% (5) | 70.56 ± 1.61% (1) | 76.07 ± 1.03% (5) | 68.24 ± 1.03% (1) |