[Retracted] DTFA-Net: Dynamic and Texture Features Fusion Attention Network for Face Antispoofing

Table 1

The network of Mag_Conv1-5.

Layer	Input size	Kernel size	Filter	Stride
Mag_Conv1	227  22  3	11  11  3	96	4
MaxPooo1	27  27  64	3  3		2
Mag_Conv2	27  27  64	5  5  64	192	1
MaxPool2	27  27  192	3  3		2
Mag_Conv3	13  13  192	3  3  192	384	1
Mag_Conv4	13  13  384	3  3  384	256	1
Mag_Conv5	13  13  256	3  3  256	256	1

VecConv1-5 and TexConv1-5 parameters are same as MagConv1-5.