Non-stationary BERT: Exploring Augmented IMU Data For Robust Human Activity Recognition

TL;DR

This paper introduces Non-stationary BERT, a lightweight model with a novel training and data augmentation approach, achieving state-of-the-art human activity recognition performance using IMU data from mobile devices.

Contribution

It presents a new lightweight BERT-based model and a data augmentation method tailored for user-specific human activity recognition with mobile phone IMU data.

Findings

Achieves state-of-the-art accuracy on multiple HAR datasets.

Demonstrates the effectiveness of the data augmentation method.

Provides a new dataset OPPOHAR for HAR research.

Abstract

Human Activity Recognition (HAR) has gained great attention from researchers due to the popularity of mobile devices and the need to observe users' daily activity data for better human-computer interaction. In this work, we collect a human activity recognition dataset called OPPOHAR consisting of phone IMU data. To facilitate the employment of HAR system in mobile phone and to achieve user-specific activity recognition, we propose a novel light-weight network called Non-stationary BERT with a two-stage training method. We also propose a simple yet effective data augmentation method to explore the deeper relationship between the accelerator and gyroscope data from the IMU. The network achieves the state-of-the-art performance testing on various activity recognition datasets and the data augmentation method demonstrates its wide applicability.