LDCformer: Incorporating Learnable Descriptive Convolution to Vision Transformer for Face Anti-Spoofing

• Published in: 2023 IEEE International Conference on Image Processing (ICIP) pp. 121 - 125
• Main Authors: Huang, Pei-Kai, Chiang, Cheng-Hsuan, Chong, Jun-Xiong, Chen, Tzu-Hsien, Ni, Hui-Yu, Hsu, Chiou-Ting
• Format: Conference Proceeding
• Language: English
• Published: IEEE 08.10.2023
• Subjects: Benchmark testing; Convolution; Convolutional codes; decoupled convolution; face anti-spoofing; Image processing; learnable descriptive convolution; Performance gain; Protocols; Representation learning; vision transformer

Face anti-spoofing (FAS) aims to counter facial presentation attacks and heavily relies on identifying live/spoof discriminative features. While vision transformer (ViT) has shown promising potential in recent FAS methods, there remains a lack of studies examining the values of incorporating local d...