AI Coach for Badminton

• Published in: 2022 3rd International Conference for Emerging Technology (INCET) pp. 1 - 7
• Main Authors: Toshniwal, Dhruv, Patil, Arpit, Vachhani, Nancy
• Format: Conference Proceeding
• Language: English
• Published: IEEE 27.05.2022
• Subjects: Badminton; Image classification and recognition; Legged locomotion; Machine Learning; Muscles; Neural Network; Neural networks; Tracking; Upper bound; Visual analytics; Web sites
• DOI: 10.1109/INCET54531.2022.9825164

In this competitive world of sports, to become an ideal player, one must maintain his nutrition and physique. Every motion involved in the sport must be efficiently carried out to employ the active muscles and eventually sustain energy depletion levels. Badminton is one sport where player movements are relatively easier to track using video analytics. The stroke movements captured through the camera can be used to examine the hand and hip coordination, placement of the leg and the varied upper bound and lower bound angles of the strokes. The prediction of a much better stance and orientation of the muscles, erroneous playing techniques, joint fatigue and several such aspects can be forecasted through an intelligent system as a recommendation. This research focuses on various techniques of neural networks to analyse the image retrieved from a badminton game. Ideal and efficient data is vastly available on the World Wide Web, and a series of data is then be taken down as per the requirements of the subjects, i.e., taking into consideration, player's body type and consecutively keeping the subjects diet and physique under observation while noting down the muscles which get activated during the execution of the said stroke.