AI-Generated Fall Data: Assessing LLMs and Diffusion Model for Wearable Fall Detection
This addresses the problem of limited fall data for wearable detection systems, particularly for elderly care, but is incremental as it builds on existing synthetic data methods.
The study tackled the scarcity of real-world fall data for training detection systems by evaluating LLMs and diffusion models for generating synthetic fall data, finding that dataset characteristics like frequency and sensor placement significantly influence effectiveness, with LLM-generated data performing best at 20Hz but showing instability at 200Hz.
Training fall detection systems is challenging due to the scarcity of real-world fall data, particularly from elderly individuals. To address this, we explore the potential of Large Language Models (LLMs) for generating synthetic fall data. This study evaluates text-to-motion (T2M, SATO, ParCo) and text-to-text models (GPT4o, GPT4, Gemini) in simulating realistic fall scenarios. We generate synthetic datasets and integrate them with four real-world baseline datasets to assess their impact on fall detection performance using a Long Short-Term Memory (LSTM) model. Additionally, we compare LLM-generated synthetic data with a diffusion-based method to evaluate their alignment with real accelerometer distributions. Results indicate that dataset characteristics significantly influence the effectiveness of synthetic data, with LLM-generated data performing best in low-frequency settings (e.g., 20Hz) while showing instability in high-frequency datasets (e.g., 200Hz). While text-to-motion models produce more realistic biomechanical data than text-to-text models, their impact on fall detection varies. Diffusion-based synthetic data demonstrates the closest alignment to real data but does not consistently enhance model performance. An ablation study further confirms that the effectiveness of synthetic data depends on sensor placement and fall representation. These findings provide insights into optimizing synthetic data generation for fall detection models.