Contrastive Left-Right Wearable Sensors (IMUs) Consistency Matching for HAR

TL;DR

This paper introduces a self-supervised learning method for human activity recognition using symmetry in wearable sensor data, improving performance over existing methods especially with limited training data.

Contribution

The paper proposes a contrastive matching approach leveraging sensor symmetry for self-supervised learning in HAR, demonstrating significant improvements over baseline methods.

Findings

Significant performance improvements over supervised and self-supervised baselines.

Effective even with limited labeled training data.

Applicable to datasets like Opportunity and MM-Fit.

Abstract

Machine learning algorithms are improving rapidly, but annotating training data remains a bottleneck for many applications. In this paper, we show how real data can be used for self-supervised learning without any transformations by taking advantage of the symmetry present in the activities. Our approach involves contrastive matching of two different sensors (left and right wrist or leg-worn IMUs) to make representations of co-occurring sensor data more similar and those of non-co-occurring sensor data more different. We test our approach on the Opportunity and MM-Fit datasets. In MM-Fit we show significant improvement over the baseline supervised and self-supervised method SimCLR, while for Opportunity there is significant improvement over the supervised baseline and slight improvement when compared to SimCLR. Moreover, our method improves supervised baselines even when using only a small amount of the data for training. Future work should explore under which conditions our method is beneficial for human activity recognition systems and other related applications.