Meta-rPPG: Remote Heart Rate Estimation Using a Transductive Meta-Learner
This addresses the challenge of reliable heart rate monitoring without contact for healthcare or fitness applications, but it is incremental as it builds on existing meta-learning and rPPG methods.
The paper tackles the problem of remote heart rate estimation using rPPG, which is sensitive to factors like skin tone and lighting, by proposing a transductive meta-learner that adapts to distributional changes during deployment, achieving state-of-the-art performance on MAHNOB-HCI and UBFC-rPPG datasets.
Remote heart rate estimation is the measurement of heart rate without any physical contact with the subject and is accomplished using remote photoplethysmography (rPPG) in this work. rPPG signals are usually collected using a video camera with a limitation of being sensitive to multiple contributing factors, e.g. variation in skin tone, lighting condition and facial structure. End-to-end supervised learning approach performs well when training data is abundant, covering a distribution that doesn't deviate too much from the distribution of testing data or during deployment. To cope with the unforeseeable distributional changes during deployment, we propose a transductive meta-learner that takes unlabeled samples during testing (deployment) for a self-supervised weight adjustment (also known as transductive inference), providing fast adaptation to the distributional changes. Using this approach, we achieve state-of-the-art performance on MAHNOB-HCI and UBFC-rPPG.