Table 4 Ablation experiments on the OBC100 dataset

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

LCE_total

Ltriplet_total

CAM

FFM

Accuracy in each session↑

  
    

0

1

2

3

4

5

6

7

8

PD↓

ΔPD

×

×

×

×

80.827

77.525

75.024

73.024

69.651

68.00

65.87

64.141

62.487

18.34

 

✓

×

×

×

80.667

77.881

75.707

73.942

70.522

68.584

66.431

65.023

63.478

17.189

+1.151

×

✓

×

×

80.364

77.153

74.85

73.27

69.918

68.0

66.102

64.8

63.022

17.342

+0.998

×

×

✓

×

80.525

76.932

74.077

73.017

69.793

67.935

65.87

64.224

62.922

17.603

+0.771

×

×

×

✓

80.37

77.153

74.632

73.086

69.827

68.338

66.114

64.262

62.801

17.569

+0.737

×

×

✓

✓

81.284

78.305

75.516

74.118

70.657

68.831

66.626

65.261

63.544

17.74

+0.6

×

✓

✓

✓

81.123

78.661

75.61

74.43

71.349

69.468

67.211

65.921

64.079

17.044

+1.296

✓

✓

✓

✓

80.994

78.102

76.67

74.405

71.15

69.571

67.259

66.089

64.523

16.471

+1.869

  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