# RARR : Robust Real-World Activity Recognition with Vibration by Scavenging Near-Surface Audio Online

## TL;DR

This paper introduces RARR, a robust activity recognition framework using vibration and near-surface audio data, capable of adapting to new environments with minimal labeled data for effective remote monitoring.

## Contribution

The work presents a scalable method that synthesizes near-surface audio data for pretraining, enabling effective activity recognition with limited real-world labeled data.

## Key findings

- Pretrained models improve activity recognition accuracy.
- Minimal fine-tuning achieves robust performance in new environments.
- Vibration-based sensing preserves privacy and enhances unobtrusiveness.

## Abstract

One in four people dementia live alone, leading family members to take on caregiving roles from a distance. Many researchers have developed remote monitoring solutions to lessen caregiving needs; however, limitations remain including privacy preserving solutions, activity recognition, and model generalizability to new users and environments. Structural vibration sensor systems are unobtrusive solutions that have been proven to accurately monitor human information, such as identification and activity recognition, in controlled settings by sensing surface vibrations generated by activities. However, when deploying in an end user's home, current solutions require a substantial amount of labeled data for accurate activity recognition. Our scalable solution adapts synthesized data from near-surface acoustic audio to pretrain a model and allows fine tuning with very limited data in order to create a robust framework for daily routine tracking.