Human-independent activity recognition of construction worker

With recent advancements in sensor and data analysis technology, multiple research on worker activity recognition through wearable sensors have been conducted to solve safety productivity problem at construction sites. However, most rely pre-trained models which require re-training of each take into account differences between workers. To alleviate this limitation, we propose a human-independent model that can adapt Our uses variational-denoising autoencoder with soft parameter sharing extract common features different activities, achieving 78.64% accuracy is higher than existing benchmark models.