Synthetic Data Generation with Few-Shot Adaptation for HAR Using Wearable Sensors via Conditional SeriesGAN

• Jonathan C. F. Silva
• Mateus C. Silva
• Vicente J. P. Amorim
• Pedro S. O. Lazaroni
• Ricardo A. R. Oliveira

Resumo

In this work, we explore the generation of synthetic data for Human Activity Recognition (HAR) using wearable sensors through a conditional version of SeriesGAN. We propose a few-shot adaptation mechanism that adjusts conditional embeddings or the final layers of the generator using a small number of genuine windows, enabling the generated data to be personalized for specific scenarios and potentially for new users in the future. We evaluate the approach using a real dataset developed by us, measuring both synthetic quality and classification performance. The results show that the conditional SeriesGAN improves classifier generalization, even in data-scarce settings, and that few-shot adaptation significantly increases accuracy when handling new scenarios.

Palavras-chave: Training, Accuracy, Scalability, Manuals, Transformers, Real-time systems, Human activity recognition, Wearable devices, Wearable sensors, Synthetic data, HAR, SPU, WPU, wearable, SeriesGAN, few-shot, synthetic data