Pose-Driven Badminton Performance Evaluation

Pose-Driven Badminton Performance Evaluation

Author Name : Mamidala Harini, Kuppili Vinay Kumar, Mallula Mukesh, Abhishek Joseph, Mr. Dharmendar K Roy

ABSTRACT In badminton, fast reflexes and quick thinking are required for various shots, each associated with peculiarities and techniques. It is a matter of learning how to hit the shuttlecock and understanding the elements that lead to successful shots. According to our research, badminton players' performance can be evaluated through posture estimation methods. This research is based on two main objectives: the first one is to develop a method that would, based on the characterization of different kinds of badminton shots, for instance, smashes, drop shots, or net shots, identify images processed using certain algorithms-the CNN algorithm for shot type classification. This should be significant as it lays the foundation for further analysis. The other goal is to determine what makes a great shot from an otherwise bad one. We run machine learning algorithms to extract salient features that appear in the successful shot samples. For instance, such features depict enhanced areas concerning the shots of novice players. This kind of study helps beginners and intermediate players as they learn new things and get opportunities to practice targeted skills.