A Probabilistic Model of Activity Recognition with Loose Clothing

TL;DR

This paper introduces a probabilistic model explaining why sensors embedded in clothing can outperform rigid sensors in activity recognition, especially over short time windows, supported by experiments.

Contribution

It presents a novel probabilistic model that accounts for improved accuracy of cloth-attached sensors in activity recognition, validated through simulations and real experiments.

Findings

Cloth-attached sensors can outperform rigid sensors in activity recognition.

The probabilistic model explains increased statistical distance in fabric sensing.

Experimental results confirm the model's predictions.

Abstract

Human activity recognition has become an attractive research area with the development of on-body wearable sensing technology. With comfortable electronic-textiles, sensors can be embedded into clothing so that it is possible to record human movement outside the laboratory for long periods. However, a long-standing issue is how to deal with motion artefacts introduced by movement of clothing with respect to the body. Surprisingly, recent empirical findings suggest that cloth-attached sensor can actually achieve higher accuracy of activity recognition than rigid-attached sensor, particularly when predicting from short time-windows. In this work, a probabilistic model is introduced in which this improved accuracy and resposiveness is explained by the increased statistical distance between movements recorded via fabric sensing. The predictions of the model are verified in simulated and real human motion capture experiments, where it is evident that this counterintuitive effect is closely captured.