Table 1 Detailed accuracy for each incremental session on the OBC100 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

80.283

74.416

69.101

64.494

60.463

56.432

53.322

50.536

48.027

32.256

+15.785

CEC25

77.605

73.712

71.908

69.389

66.097

64.0

61.244

60.053

59.354

18.251

+1.78

CLOM46

84.03

79.81

75.39

71.39

68.38

66.07

63.61

62.52

60.76

23.27

+6.799

S3C26

75.709

72.355

69.59

67.43

65.106

62.909

60.489

60.049

58.667

17.042

+0.571

FACT7

81.704

77.593

75.888

73.924

70.639

68.571

66.492

64.855

63.221

18.483

+2.012

SAVC24

71.79

67.966

64.863

63.615

60.929

58.935

57.579

56.284

53.679

18.111

+1.67

Ours

80.994

78.102

76.67

74.405

71.15

69.571

67.259

66.089

64.523

16.471

 
  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