Human Activity Recognition from Wearable Sensor Data Using Self-Attention

TL;DR

This paper introduces a self-attention based neural network for human activity recognition from wearable sensor data, effectively capturing spatial and temporal dependencies without recurrent structures, leading to improved accuracy.

Contribution

The study presents a novel self-attention model that outperforms existing methods in HAR tasks and highlights the importance of sensor modality and placement through attention maps.

Findings

Significant accuracy improvements over state-of-the-art models.

Effective capture of sensor importance and placement.

Robust performance across multiple datasets.

Abstract

Human Activity Recognition from body-worn sensor data poses an inherent challenge in capturing spatial and temporal dependencies of time-series signals. In this regard, the existing recurrent or convolutional or their hybrid models for activity recognition struggle to capture spatio-temporal context from the feature space of sensor reading sequence. To address this complex problem, we propose a self-attention based neural network model that foregoes recurrent architectures and utilizes different types of attention mechanisms to generate higher dimensional feature representation used for classification. We performed extensive experiments on four popular publicly available HAR datasets: PAMAP2, Opportunity, Skoda and USC-HAD. Our model achieve significant performance improvement over recent state-of-the-art models in both benchmark test subjects and Leave-one-subject-out evaluation. We also observe that the sensor attention maps produced by our model is able capture the importance of the modality and placement of the sensors in predicting the different activity classes.