计算机科学
欺骗攻击
模态(人机交互)
人工智能
面子(社会学概念)
模式
保险丝(电气)
任务(项目管理)
面部识别系统
无线传感器网络
基线(sea)
计算机视觉
模式识别(心理学)
计算机安全
计算机网络
社会科学
海洋学
管理
社会学
地质学
电气工程
经济
工程类
作者
Pengchao Deng,Chenyang Ge,Hao Wei,Yu-An Sun,Xin Qiao
标识
DOI:10.1016/j.engappai.2023.107600
摘要
Multimodal face anti-spoofing systems adopt multiple sensor modalities, such as infrared, color, depth, and thermal, to distinguish between living and spoofing faces via complementary spoofing clues from each modality. One challenge is that when the multimodal face anti-spoofing system is placed in different environments, the sensor setup may not be unified, causing a certain sensor to be unavailable. To alleviate this issue, a two-stream face anti-spoofing method is proposed. The first stream focuses on extracting primary features from an available sensor by a baseline network. The second stream employs a multimodal contrastive learning strategy to acquire modality-agnostic and task-specific representations from another deployed sensor. Furthermore, a master–slave modulation fusion block is designed to effectively fuse features from the two streams. Experiments conducted on three public multimodal databases show the superior performance of the proposed method.
科研通智能强力驱动
Strongly Powered by AbleSci AI