Deep - learning based Human Activity Detection Model

In computer vision, human activity detection is a challenging task with applications spanning from man-machine interaction to intelligence and security. Human action detection is the field of computational science and engineering that develops methods and systems for automatically identifying or classifying human activities through the use of deep learning and machine learning. This study reviews many cutting-edge approaches to human action detection (HAD) based on deep learning and machine learning. Subsequently, an interactive framework based on 3D skeletal data was created to distinguish different human movements. In this study, a media pipe was utilized to detect the different movements of the human body utilizing three popular classifiers including logistic regression random forest and gradient boosting classifier. The proposed framework performs better than numerous cutting-edge techniques, such as CNN and RNN achieving a classification report indicating accuracy as 97.33%.