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

Figure 4

Comparison between [20] and ours. The improved algorithm we proposed performs better than that in [20]. (a) Original images that cannot detect face by HOG; (b) images processed in [20] and the detection result; (c) images processed by ours and the detection result.

(a)

(b)

(c)