Bassel Emad, Omar Atef, , Yahya Shams, Ahmed ElKerdany
8th October 2019
Smart coaching in martial arts is one of the recent research areas in Human Motion Analysis. Numerous moves are performed incorrectly during the performance. In this project we offer a system that will record the Players’ movements using IR (Infrared) camera sensor, store the data in a database, pre-process the data, classify the data using F-DTW (Fast Dynamic Time Warping) and then show the users an accurate report that contains every movement the player had done, their mistakes and how to improve their performance the next time. This approach focuses on the Karate Kata 1 (Hein Shodan) moves. The system has reached an accuracy of 93.65% in classifying the moves and an incorrect version of the moves that involved a common mistake.
Karate moves are combination of successive moves. Kids nowadays may find difficulties learning those moves at a young age, since the training may consist of a large number of players, the trainer himself may not be able to focus on every detail of every player’s move, which may result in taking more time for learning the move or to be precious to master it or it may lead to learning that move in a wrong way from the beginning. The main goal of this project is to capture the moves of the players, analyse those moves and give the players a feedback to enhance their technique or alert them if they are playing in a wrong way.
Despite the importance of the sports field and improving it, not many people focus on working in the Karate sport, so as a team we saw a good opportunity to continue researching in the Karate sport, by improving and adding some features that might make a difference. We also found that kids playing Karate at the beginning don’t have the attention needed from the coach who should focus on the little mistakes made by the kids.
1.3 Problem Statement
There is the problem of the real-time feedback, giving the player a feedback on his/her moves whether it was right or not in a couple of minutes is essential after he/she finishes the move, the feedback should have high-accuracy. The feedback contains tips on how to execute the move correctly. Moreover, each player has different body proportions than the other players.