Towards LLM-Powered Ambient Sensor Based Multi-Person Human Activity Recognition

TL;DR

This paper introduces LAHAR, a novel LLM-based framework for multi-person human activity recognition that leverages prompt engineering to improve accuracy and robustness in complex environments.

Contribution

The paper presents LAHAR, a new system that applies large language models and prompt techniques to enhance multi-person HAR, addressing data scarcity and generalization issues.

Findings

LAHAR achieves comparable accuracy to state-of-the-art methods.

The approach maintains robustness in multi-person scenarios.

Validated on the ARAS dataset with promising results.

Abstract

Human Activity Recognition (HAR) is one of the central problems in fields such as healthcare, elderly care, and security at home. However, traditional HAR approaches face challenges including data scarcity, difficulties in model generalization, and the complexity of recognizing activities in multi-person scenarios. This paper proposes a system framework called LAHAR, based on large language models. Utilizing prompt engineering techniques, LAHAR addresses HAR in multi-person scenarios by enabling subject separation and action-level descriptions of events occurring in the environment. We validated our approach on the ARAS dataset, and the results demonstrate that LAHAR achieves comparable accuracy to the state-of-the-art method at higher resolutions and maintains robustness in multi-person scenarios.