Intelligent Elderly Monitoring System Using Computer Vision

This project presents an intelligent monitoring system that uses computer vision technologies to improve the care of older adults. Various tests were carried out to demonstrate the accuracy and effectiveness of the selected machine learning models, as well as comparisons between the training data used, including different training/test split divisions to determine the best configuration. In addition, a proprietary dataset with 300 videos was created and compared with a subset of NTU RGB+D. The results showed high accuracy and effectiveness in activity detection and tracking.