tech
Google: Detecting Heart Rate Through Smartphone Camera AI
Google researchers developed technology using smartphone cameras to monitor heart rate passively during everyday use with medical-grade accuracy
TL;DR
- Google researchers have created a method to passively monitor heart rate using smartphone cameras and AI.
- The technology analyzes facial video captured during normal smartphone use to estimate heart rate.
- It achieves medical-grade accuracy with a mean absolute percentage error of less than 10%, meeting industry standards across all skin tones.
- This is the first large-scale demonstration of passive heart rate monitoring during everyday smartphone use.
- The system uses temporal shift convolutional neural networks and has been trained on a diverse dataset of over 350,000 video clips.
- Google plans to make its data and modeling resources available to qualified researchers.