Accelerometer based Activity Classification with Variational Inference on Sticky HDP-SLDS

TL;DR

This paper introduces a fast variational inference method for an unsupervised non-parametric model to classify indoor human activities using smartphone accelerometer data, outperforming traditional methods in speed and accuracy.

Contribution

The paper presents a novel variational inference approach for sticky HDP-SLDS, enabling efficient unsupervised activity classification from accelerometer data.

Findings

The proposed method accurately differentiates indoor activities.

Variational inference is ten times faster than Gibbs sampling.

The approach outperforms Hidden Markov Models in classification accuracy.

Abstract

As part of daily monitoring of human activities, wearable sensors and devices are becoming increasingly popular sources of data. With the advent of smartphones equipped with acceloremeter, gyroscope and camera; it is now possible to develop activity classification platforms everyone can use conveniently. In this paper, we propose a fast inference method for an unsupervised non-parametric time series model namely variational inference for sticky HDP-SLDS(Hierarchical Dirichlet Process Switching Linear Dynamical System). We show that the proposed algorithm can differentiate various indoor activities such as sitting, walking, turning, going up/down the stairs and taking the elevator using only the acceloremeter of an Android smartphone Samsung Galaxy S4. We used the front camera of the smartphone to annotate activity types precisely. We compared the proposed method with Hidden Markov Models with Gaussian emission probabilities on a dataset of 10 subjects. We showed that the efficacy of the stickiness property. We further compared the variational inference to the Gibbs sampler on the same model and show that variational inference is faster in one order of magnitude.