MobilePhys: Personalized Mobile Camera-Based Contactless Physiological Sensing
2022年3月1日·
,
,
,
,
,
·
0 分钟阅读时长
Xin Liu
Yuntao Wang
Sinan Xie
Xiaoyu Zhang
Zixian Ma
Daniel McDuff
Shwetak Patel
摘要
基于摄像头的非接触式光电容积描记（rPPG）是一类用于非接触生理测量的流行技术。当前最先进的神经网络模型通常采用监督学习方式，使用配有金标准生理测量的视频进行训练。然而，这些模型在域外样本（即与训练集不同的视频）上往往泛化性能较差。个性化模型有助于提升泛化能力，但许多个性化技术仍需要一定的金标准数据。为缓解这一依赖，本文提出了一种新型移动传感系统MobilePhys——首个移动个性化远程生理传感系统，该系统利用智能手机的前后双摄像头生成高质量的自监督标签，用于训练个性化的非接触式摄像头PPG模型。为评估MobilePhys的鲁棒性，我们开展了一项包含39名参与者的用户研究，参与者在不同移动设备、光照条件/强度、运动任务和肤色条件下完成了一系列任务。研究结果表明，MobilePhys显著优于最先进的设备端监督训练和少样本适应方法。通过大规模用户研究，我们进一步考察了MobilePhys在复杂真实场景中的表现。我们预期，通过所提出的双摄像头移动传感系统生成的校准或个性化非接触式PPG模型，将为智能镜面、健身及移动健康应用等众多未来应用开启新的可能。
类型
出版物
Proc. ACM Interact. Mob. Wearable Ubiquitous Technol.
Authors
Associate Professor (Research Track)
Yuntao Wang’s research centers on physiobehavioral computing and
intelligent interaction for mobile and wearable systems. His work
focuses on (1) developing robust, efficient sensing that performs
reliably on mainstream devices, (2) extracting spatiotemporal patterns
from multimodal signals to infer interaction intent by leveraging
natural behavioral correlations, and (3) designing edge-efficient interfaces
that deliver high performance on mobile and wearable platforms.
He has published 90+ papers, received 10 international conference awards,
and holds 30+ granted patents. His contributions have been
recognized with honors including the Wu Wenjun AI Outstanding Youth
Award (2024), the CAST Young Elite Scientists Sponsorship Program (2022),
the Qinghai High-Level Innovation & Entrepreneurship Leading Talent (2024),
and the First Prize of the China Electronics Institute Science & Technology Award (2019).