Personalized Fall Detection by Balancing Data with Selective Feedback Using Contrastive Learning

TL;DR

This paper introduces a personalized fall detection method that balances user feedback data using contrastive learning, significantly improving detection accuracy across different retraining strategies in real-world tests.

Contribution

It presents a novel framework combining semi-supervised clustering and contrastive learning to select and balance informative feedback samples for personalized fall detection.

Findings

TFS achieves up to 25% performance improvement.

FSL achieves 7% improvement, showing effectiveness.

Framework adapts well across different learning paradigms.

Abstract

Personalized fall detection models can significantly improve accuracy by adapting to individual motion patterns, yet their effectiveness is often limited by the scarcity of real-world fall data and the dominance of non-fall feedback samples. This imbalance biases the model toward routine activities and weakens its sensitivity to true fall events. To address this challenge, we propose a personalization framework that combines semi-supervised clustering with contrastive learning to identify and balance the most informative user feedback samples. The framework is evaluated under three retraining strategies, including Training from Scratch (TFS), Transfer Learning (TL), and Few-Shot Learning (FSL), to assess adaptability across learning paradigms. Real-time experiments with ten participants show that the TFS approach achieves the highest performance, with up to a 25% improvement over the baseline, while FSL achieves the second-highest performance with a 7% improvement, demonstrating the effectiveness of selective personalization for real-world deployment.