Latest Deep Learning Techniques for Fall Detection in Monitoring Real-Time Video Data

Escorted by the latest improvements in technology, real-time monitoring of video data plays a rising supreme role in our lives. It has already had a significant influence on various areas of indoor and outdoor activities. To correctly classify the suspicious event as fall detection in monitoring real-time video data, latest deep learning techniques are in the demand of computer vision. Detecting correct suspicious events is highly famous and demanding in monitoring real-time video. In the paper, real-time events are considered for monitoring a person living in an indoor environment and performing their own daily activity in a normal manner. Suspicious events are considered as detecting the falling of a person. As the falling of a person is not a normal event, it is considered a suspicious event in real time. Although different techniques have been developed to efficiently classify suspicious events of fall detection, all deep learning techniques are the most famous. A brief overview of various machine learning and deep learning techniques applied by various researchers for detecting suspicious events, such as fall detection in the monitoring of real-time video data, has been considered. The paper also exposes different affairs of machine learning and deep learning techniques that may encourage intended researchers to work in the domain of monitoring video data.