Skip to main content
. 2021 Jan 18;2021:6235319. doi: 10.1155/2021/6235319

Table 3.

Calculation efficiency of different models.

Model Memory space (GB) Training time (h) Parameters
FCN 8.45 12.6 63,254,618
SegNet 6.13 10.3 56,421,514
U-Net 5.42 8.6 48,163,254
DeepLab 7.13 7.3 52,618,315
PSPNet 4.26 7.8 45,126,257
DS-CNNs 3.42 4.5 35,183,326