<ul><li>MobiVital is a self-supervised approach for improving the quality of respiration waveforms obtained from UWB radar data.</li><li>It combines a self-supervised autoregressive model for breathing waveform extraction with a biology-informed algorithm to detect and correct waveform inversions.</li><li>A 12-person, 24-hour UWB radar vital signal dataset with time-synchronized ground truth is released for reproducible research efforts.</li><li>The results show an increase in fidelity to the ground truth, benefiting downstream tasks like respiration rate estimation.</li></ul>

MobiVital: Self-supervised Time-series Quality Estimation for Contactless Respiration Monitoring Using UWB Radar

Discover more