Skip to content

This repository helps to track the exercise movements of sports player using poseestimation. Using this mediapipe poseestimation, we can also calculate score and angle calculation w.r.t. video or live streaming. I've implemented the same for physiotherapy exercises and yoga.

Notifications You must be signed in to change notification settings

jhanvi-29/Exercise-Instructor-using-AI

Repository files navigation

Exercise Instructor using AI

This repository contains code and resources for an AI-powered exercise instructor that analyzes exercise videos, extracts key points, calculates angles, and evaluates performance in real-time.

Project Structure

  • KeyPointExtraction_AngleCalculation.ipynb: Jupyter notebook for extracting key points from exercise videos and calculating angles.
  • ROI_frames.ipynb: Jupyter notebook for identifying and processing regions of interest (ROI) in video frames.
  • ROI_from_video.ipynb: Jupyter notebook for extracting ROI from videos.
  • RT-LessScore.ipynb: Jupyter notebook for real-time scoring based on extracted key points and angles.
  • videoLateral.mp4: Sample exercise video used for testing and demonstration.

Getting Started

Prerequisites

  • Python 3.x
  • Jupyter Notebook
  • OpenCV
  • NumPy
  • Pandas
  • Other dependencies listed in the respective notebooks

Installation

  1. Clone the repository:
    git clone https://github.com/yourusername/Exercise-Instructor-using-AI.git
    cd Exercise-Instructor-using-AI
  2. pip install -r requirements.txt

About

This repository helps to track the exercise movements of sports player using poseestimation. Using this mediapipe poseestimation, we can also calculate score and angle calculation w.r.t. video or live streaming. I've implemented the same for physiotherapy exercises and yoga.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published