Stroke Impact Frame Detection in Badminton

Project Code :TCMAPY2000

Objective

The objective of this project is to develop a real-time badminton stroke classification system using deep learning techniques, specifically CNN for spatial feature extraction and GRU for temporal analysis of video data.

Abstract

This project focuses on detecting stroke impact frames in badminton using video analysis. The goal is to classify various badminton strokes, such as backhand drive, forehand clear, forehand net shot, and others, from video frames. The project employs convolutional neural networks (CNN) combined with GRU (Gated Recurrent Units) to extract meaningful features from video frames and classify the actions. Data is gathered from Kaggle, and the models are trained and evaluated using custom-built training pipelines. The system is deployed with a Flask backend and an interactive frontend using HTML, CSS, and JavaScript. The project also utilizes a MySQL database to manage users and store prediction history.

Keywords: Stroke impact detection, badminton, CNN, GRU, video classification, deep learning, Kaggle dataset, Flask, prediction, action recognition.

NOTE: Without the concern of our team, please don't submit to the college. This Abstract varies based on student requirements.