Table 4 Statistics on evaluation metrics for the classification results of the five programs in the sensaturban dataset (%).

From: An aerial point cloud classification using point transformer via multi-feature fusion

Features

Acc

Ground

Vegetation

Building

Wall

Bridge

Parking

Rail

Traffic road

Street furniture

Car

Footpath

Bike

Water

mIoU

mAc c

OA

feat6

91.17

97.88

94.67

89.38

67.78

64.91

0

96.65

66.25

88.67

73.90

0

87.79

58.35

70.67

90.61

feat7

94.35

98.79

97.58

86.73

70.77

79.41

0

90.99

43.05

90.51

85.10

0

98.02

59.23

71.94

91.23

feat8

96.22

98.38

97.92

83.76

65.43

92.09

0

86.35

74.17

87.67

68.20

0

91.77

61.88

72.07

92.94

feat9

96.11

99.42

98.23

81.09

91.22

90.08

0

89.14

73.73

91.73

89.03

0

56.61

62.03

73.56

93.20

feat10

96.17

98.22

97.46

80.16

94.19

92.09

0

72.45

80.92

93.24

89.70

0

79.66

63.52

74.94

94.18