Skip to main content

Table 4 Evaluation with dataset A

From: A hybrid framework combining background subtraction and deep neural networks for rapid person detection

Input size

Precision

Recall

Accuracy

Person

Non-person

Person

Non-person

(a) Full-set

 32 × 32

0.80

0.83

0.82

0.81

0.81

 64 × 64

0.83

0.83

0.81

0.84

0.83

 128 × 128

0.83

0.85

0.85

0.84

0.84

(b) Part-set

 32 × 32

0.82

0.82

0.82

0.81

0.82

 64 × 64

0.87

0.82

0.83

0.86

0.84

 128 × 128

0.86

0.84

0.84

0.86

0.85