Human Activity Recognition on wrist-worn accelerometers using self-supervised neural networks
This work addresses the challenge of remote health monitoring by enabling automated and accurate activity recognition, which is incremental as it builds on existing self-supervised and segmentation techniques for a specific domain.
The paper tackled the problem of human activity recognition from wrist-worn accelerometers by developing a self-supervised learning method to create robust representations that generalize across devices and subjects, achieving strong accuracy on benchmark datasets with very few labels and boosting performance on continuous real-life data through a segmentation algorithm.
Measures of Activity of Daily Living (ADL) are an important indicator of overall health but difficult to measure in-clinic. Automated and accurate human activity recognition (HAR) using wrist-worn accelerometers enables practical and cost efficient remote monitoring of ADL. Key obstacles in developing high quality HAR is the lack of large labeled datasets and the performance loss when applying models trained on small curated datasets to the continuous stream of heterogeneous data in real-life. In this work we design a self-supervised learning paradigm to create a robust representation of accelerometer data that can generalize across devices and subjects. We demonstrate that this representation can separate activities of daily living and achieve strong HAR accuracy (on multiple benchmark datasets) using very few labels. We also propose a segmentation algorithm which can identify segments of salient activity and boost HAR accuracy on continuous real-life data.