Table 4.

SC-FSL vs. some state-of-the-art FSL.

Algorithms	Accuracy (%)
	5-way 1-shot	5-way 5-shot
ProtoNet	75.32 ± 0.80	89.70 ± 0.51
MatchingNet	76.80 ± 0.81	87.85 ± 0.56
RelationNet	74.71 ± 0.83	88.90 ± 0.40
NegMargin	72.40 ± 0.80	90.78 ± 0.47
MetaBaseline	70.07 ± 0.81	87.02 ± 0.51
MAML	69.97 ± 0.96	86.04 ± 0.56
FEAT	74.23 ± 0.03	88.01 ± 0.03
MELR	74.90 ± 0.75	89.02 ± 0.13
DeepEMD	73.87 ± 0.07	88.56 ± 0.06
SC-FSL(ours)	78.55 ± 0.81	92.90 ± 0.47

The boldface is the best result and the underline the second-ranked result.