Table 3 Comparison of the OBI classification accuracy of different methods on OBIR-10 (%)

From: OBI-CMF: Self-supervised learning with contrastive masked frequency modeling for oracle bone inscription recognition

Method

Supervised

Top-1

Top-5

GoogLeNet49

T

99.43

99.95

DenseNet19

T

99.46

99.95

ResNet14

T

99.35

99.96

VGG13

T

99.32

99.95

ViT28

T

98.53

99.96

SimSiam32

F

98.64

99.95

MoCov333

F

98.88

99.98

MAE26

F

99.43

99.97

MFM30

F

99.41

99.97

Ours

F

99.57

99.98

  1. Values in bold indicate optimal results.
  2. T/F represents True/False.
Back to article page