Table 2 Detailed accuracy for each incremental session on the HWAYI100 dataset

From: DFS: Dual-branch forward-looking simulation network for incremental learning of ancient Chinese characters

Method

Accuracy in each session↑

  
 

0

1

2

3

4

5

6

7

8

PD↓

ΔPD

Finetune

96.677

92.463

87.326

82.73

74.851

69.347

64.598

61.242

58.945

37.732

+28.887

CEC25

93.71

92.921

91.501

90.959

89.787

88.356

87.853

85.70

84.684

9.026

+0.181

CLOM46

96.32

95.02

92.99

92.51

91.90

89.37

88.21

87.89

86.27

10.05

+1.025

S3C26

91.397

89.363

88.026

85.882

84.366

82.068

80.883

79.608

78.641

12.756

+3.911

FACT7

96.851

95.14

93.692

91.306

91.353

89.08

88.243

88.114

87.057

9.794

+0.949

SAVC24

89.686

88.413

86.495

83.55

82.753

81.227

79.504

78.852

77.484

12.202

+3.357

Ours

96.851

95.417

94.047

91.631

91.864

90.159

89.142

88.787

88.006

8.845

 
  1. The bold values in the last row represent the recognition accuracy of our proposed method in the incremental stage, and the bold values in the last column represent the improved accuracy of our proposed method in the forgetting rate indicator compared with other methods.
Back to article page