计算机科学
人工智能
计算机视觉
RGB颜色模型
卷积神经网络
呼吸
代表(政治)
模式识别(心理学)
运动(物理)
计量系统
深度学习
反射(计算机编程)
主管(地质)
特征提取
机制(生物学)
作者
Weixuan Chen,Daniel McDuff
标识
DOI:10.48550/arxiv.1805.07888
摘要
Non-contact video-based physiological measurement has many applications in health care and human-computer interaction. Practical applications require measurements to be accurate even in the presence of large head rotations. We propose the first end-to-end system for video-based measurement of heart and breathing rate using a deep convolutional network. The system features a new motion representation based on a skin reflection model and a new attention mechanism using appearance information to guide motion estimation, both of which enable robust measurement under heterogeneous lighting and major motions. Our approach significantly outperforms all current state-of-the-art methods on both RGB and infrared video datasets. Furthermore, it allows spatial-temporal distributions of physiological signals to be visualized via the attention mechanism.
科研通智能强力驱动
Strongly Powered by AbleSci AI