Towards a New Multi-tasking Learning Approach for Human Fall Detection

Abstract

Many fall detection systems are being used to provide real-time responses to fall occurrences. Automated fall detection is challenging because it requires near perfect accuracy to be clinically acceptable. Recent research has tried to improve the accuracy along with reducing the high rate of false positives. Nevertheless, there are still limitations in terms of having efficient learning approaches and proper datasets to train. To improve the accuracy, one approach is to include non-fall data from public datasets as negative examples to train the deep learning model. However, this approach could increase the imbalance of the training set. In this paper, we propose a multi-task deep learning model to tackle this problem. We divide datasets into multiple training sets for multiple tasks, and we prove this approach gives better results than a single-task model trained on all datasets. Many experiments are conducted to find the best combination of tasks for multi-task model training for fall detection.