Skip to main content

Advances, Systems and Applications

Table 1 Comparison of different models on Landsat-8 Dataset

From: PMNet: a multi-branch and multi-scale semantic segmentation approach to water extraction from high-resolution remote sensing images with edge-cloud computing

Model Name

Recall\(\uparrow\)

F1-score\(\uparrow\)

Params\(\downarrow\)

GFLOPs\(\downarrow\)

Memory\(\downarrow\)

FCN

0.8269

0.8527

15.3M

80.5

571.01MB

DeepLabV3

0.7755

0.8092

39.6M

25.5

578.01MB

BiseNetV1

0.7902

0.8347

12.7M

13.02

228.42MB

DDRNet

0.8113

0.8502

5.6M

4.6

115.51MB

SPNet

0.8760

0.8941

39.2M

158.9

1703.97MB

PMNet

0.8963

0.9022

14M

14.6

338.2MB

  1. Bold is optimal, underline is suboptimal