Table 2 Comparison on the DeepGlobe dataset.

From: Spatially adaptive interaction network for semantic segmentation of high-resolution remote sensing images

Method

Barren

Rangeland

Forest

Water

Urban

Agriculture

mIoU (%)

Mem (GB)

ICNet72

35.2

31.3

24.2

42.1

24.3

69.2

37.7

7.36

FCN15

36.1

36.4

35.5

38.9

26.2

72.3

40.9

6.14

U-Net18

34.1

33.5

28.2

45.1

26.1

71.7

39.8

6.89

PSPNet23

39.3

36.9

53.6

64.2

48.2

74.7

52.8

7.62

SegNet73

43.1

38.2

54.8

67.8

53.1

75.2

55.4

5.67

DeepLabv3+26

44.1

42.2

56.3

68.4

54.1

77.3

57.1

6.28

UHRSNet12

81.2

61.3

72.3

70.1

78.9

68.4

72.0

5.99

MBNet13

64.1

87.3

41.5

80.6

83.1

78.9

72.6

6.17

HRNet74

75.2

84.5

35.0

76.7

82.4

68.5

70.9

2.76

Mask2Former38

78.1

87.0

45.8

86.0

87.1

74.3

76.4

8.29

Baseline11

63.6

38.6

79.8

82.6

78.1

86.8

71.6

1.40

SAINet

86.5

46.1

83.1

84.4

79.2

87.3

77.8

1.87

  1. Significant values are in bold.