A Framework for Analyzing the Exercise and Athletic Activities

Exercise plays an important role in maintaining the health of a person. But doing exercise is not enough until it is done in a proper manner. Therefore, in this paper, we have proposed a framework for analyzing different exercise and athletic activities efficiently. The proposed approach is based on the Human Activity Recognition (HAR) which has been used for analyzing different activities related to exercising. In this paper, analysis has been carried out using the neural network classifier and has been carried out on a number of videos of different exercises. The videos are converted into frames for the methodical analysis of the videos. This framework is able to efficiently identify and distinguish the different exercises like flic flac, running, push up, pull up, sit up, somersault and the accuracy obtained is 88.53%.