VitalLens: Take A Vital Selfie
This provides a convenient, non-invasive tool for health monitoring via smartphones, but it is incremental as it builds on existing computer vision methods for vital sign estimation.
The authors tackled the problem of estimating vital signs like heart rate and respiration rate from selfie videos in real time, achieving mean absolute errors of 0.71 bpm for heart rate and 0.76 bpm for respiratory rate on a dataset of 289 participants.
This report introduces VitalLens, an app that estimates vital signs such as heart rate and respiration rate from selfie video in real time. VitalLens uses a computer vision model trained on a diverse dataset of video and physiological sensor data. We benchmark performance on several diverse datasets, including VV-Medium, which consists of 289 unique participants. VitalLens outperforms several existing methods including POS and MTTS-CAN on all datasets while maintaining a fast inference speed. On VV-Medium, VitalLens achieves mean absolute errors of 0.71 bpm for heart rate estimation, and 0.76 bpm for respiratory rate estimation.