Using Deep Learning to Recognize Human Activities from Binary Sensor Data

What is it about?

This paper presents a methodology using Deep Neural Networks (DNNs) for Human Activity Recognition (HAR) using binary sensor data. It introduces a technique that converts binary sensor data into an image map grid, which is then transformed into a video stream to capture spatial and temporal features. The experimental results show similar performance to the state-of-the-art, demonstrating the effectiveness of this approach.

Featured Image

Photo by Skiathos Greece on Unsplash

Photo by Skiathos Greece on Unsplash

Why is it important?

This method can be applied to Human Action Recognition in Assisted Living for elderly.