Skip to main content

Table 2 Noise suppression using the flipover technique

From: Flipover outperforms dropout in deep learning

Model

Method

ACC Org (%)

ACC Gaussian (%)

ACC Poisson (%)

ACC salt (%)

Small CNN [17]

Vanilla

98.53

58.31

/

50.49

Dropout

98.60

61.50

/

61.50

Flipover

98.00

68.71

/

69.62

ResNet18 [4]

Vanilla

93.63

41.27

46.92

46.71

Dropout

93.31

41.60

43.03

48.00

Flipover

92.37

46.50

49.85

53.49

  1. Note: ACC Org stands for the accuracy on the original test set; ACC Gaussian stands for the accuracy on dataset with Gaussian noise; ACC Poisson stands for the accuracy on dataset with Poisson noise; ACC salt stands for the accuracy on dataset with salt-and-pepper noise