Table 5 The influence of different sub-block quantities in LG-FEM on model performance.

From: Multi-feature enhancement fusion network for remote sensing image semantic segmentation

\(\alpha\)

mIoU (%)

Params (M)

FLOPS (G)

1

84.35

29.85

52.52

2

84.46

29.85

53.13

4

84.61

29.85

53.22

6

84.45

29.85

53.40

Back to article page