Skip to content

anandkrsah/Exercise-Posture-Correction

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

1 Commit
 
 
 
 
 
 
 
 

Repository files navigation

Real-Time AI-Based Exercise Evaluation Web App

Project Overview

This web application provides real-time fitness exercise evaluation using MediaPipe Pose Landmarker and Three.js visualization. The app analyzes user posture through webcam input and provides immediate feedback on proper form for squats and push-ups.

Features

  • Real-time pose detection using MediaPipe
  • Support for two exercises: Squats and Push-Ups
  • Rule-based posture evaluation for different exercise phases
  • Visual feedback with color-coded keypoints (red for incorrect, green for correct)
  • Three.js annotations highlighting posture issues and explaining corrections
  • Mobile-responsive design for use on various devices

Technologies Used

  • HTML5, CSS3, JavaScript
  • MediaPipe Pose Landmarker for real-time pose tracking
  • Three.js for 3D visualizations and annotations
  • Responsive design principles for cross-device compatibility

How to Use

  1. Allow camera access when prompted
  2. Select an exercise (Squats or Push-Ups) using the buttons
  3. Position yourself in the camera view with your full body visible
  4. Follow the on-screen instructions and feedback
  5. Correct your form based on the visual indicators provided

Setup Instructions

  1. Clone this repository
  2. Open index.html in your browser
  3. Alternatively, visit the live demo at [your-demo-link]

Project Structure

  • index.html: Main HTML structure and layout
  • css/style.css: Styling for the application
  • js/script.js: Core application logic including:
    • MediaPipe integration
    • Exercise detection algorithms
    • Posture evaluation rules
    • Three.js visualization setup

Implementation Details

This application implements rule-based posture evaluation logic that:

  • Breaks exercises into phases (e.g., descent and ascent for squats)
  • Monitors key joint angles and positions
  • Provides specific feedback for common form issues
  • Uses visual indicators to guide proper technique

Future Enhancements

  • Additional exercise types (lunges, planks)
  • User profiles to track progress
  • More detailed analytics on form improvement over time
  • Custom exercise programming options

Created By

Rabi Kumar Mishra - Submitted as part of the Realfy Oasis Full Stack Intern technical assignment

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors