Face presentation attack detection via content-adaptive spatial operators

Abstract

Face presentation attack detection (FacePAD) is critical for securing facial authentication against print, replay, and mask-based spoofing. This paper proposes CASO-PAD, an RGB-only, single-frame model that enhances MobileNetV3 with content-adaptive spatial operators (involution) to better capture localized spoof cues. Unlike spatially shared convolution kernels, the integrated grouped involution operator generates location-specific kernels conditioned on the input and shares each kernel within a channel group, improving spatial selectivity with limited computational overhead. CASO-PAD remains lightweight (3.6M parameters; 0.256 GFLOPs at \(256\times 256\) ) and is trained end-to-end using a standard binary cross-entropy objective. Extensive experiments on Replay-Attack, Replay-Mobile, ROSE-Youtu, and OULU-NPU demonstrate strong performance, achieving 100/100/98.9/99.7% test accuracy, AUC of 1.00/1.00/0.9995/0.9999, and HTER of 0.00/0.00/0.82/0.44%, respectively. On the large-scale SiW-Mv2 Protocol-1 benchmark, CASO-PAD attains 95.45% accuracy with 3.11% HTER and 3.13% EER, indicating strong performance under the evaluated protocol. Ablation studies show that placing the adaptive operator near the network head and using moderate group sharing yields the best accuracy–efficiency balance. Overall, CASO-PAD provides a lightweight RGB-only design for edge-oriented FacePAD under the evaluated settings.