Searching Central Difference Convolutional Networks for Face Anti-spoofing

Tóm tắt báo cáo

Face anti-spoofing (FAS) plays a vital role in face recognition systems. Here the authors propose a novel operation called Central Difference Convolution (CDC), which is able to capture intrinsic detailed patterns via aggregating both intensity and gradient information. A network built with CDC, called the Central Difference Convolutional Network (CDCN), is able to out-performed its counterpart built with vanilla convolution. Furthermore, Neural Architecture Search (NAS) is utilized to discover a more powerful network structure (CDCN++).