PoseTrack-Real-Time-Pose-Estimation

🏋️‍♂️ GymPose AI - Real-Time Pose Estimation & Joint Analysis

A cutting-edge fitness application that uses AI-powered computer vision to provide real-time pose estimation, form analysis, and personalized coaching feedback for gym exercises.

✨ Features

Real-Time Pose Detection: Advanced computer vision using OpenCV and MediaPipe
Form Analysis: Instant feedback on exercise technique with angle calculations
Rep Counting: Automatic repetition and set tracking
Progress Analytics: Detailed workout history and performance metrics
Multi-Exercise Support: Knee bends, squats, deadlifts, lunges, and more
Beautiful UI/UX: Modern React frontend with smooth animations
Secure Backend: Java Spring Boot with JWT authentication
Database: SQLite/PostgreSQL with comprehensive data models

🏗️ Architecture

GymPose AI/
├── backend/                 # Java Spring Boot Backend
│   ├── src/main/java/
│   │   └── com/gympose/
│   │       ├── controller/  # REST API endpoints
│   │       ├── service/     # Business logic & pose estimation
│   │       ├── model/       # Database entities
│   │       ├── dto/         # Data transfer objects
│   │       └── config/      # Security & configuration
│   └── pom.xml             # Maven dependencies
├── frontend/               # React.js Frontend
│   ├── src/
│   │   ├── components/     # Reusable UI components
│   │   ├── pages/          # Application pages
│   │   ├── contexts/       # React contexts
│   │   └── services/       # API integration
│   └── package.json        # Node.js dependencies
└── README.md               # This file

🚀 Quick Start

Prerequisites

Java 17+ (for backend)
Node.js 16+ (for frontend)
Maven 3.6+ (for Java dependencies)
Camera-enabled device (for pose detection)

Backend Setup (Java)

Navigate to backend directory:
```
cd backend
```
Install dependencies:
```
mvn clean install
```
Run the application:
```
mvn spring-boot:run
```
Access the API:
- Backend: http://localhost:8080/api
- H2 Database Console: http://localhost:8080/h2-console
- API Documentation: http://localhost:8080/swagger-ui.html

Frontend Setup (React)

Navigate to frontend directory:
```
cd frontend
```
Install dependencies:
```
npm install
```
Start development server:
```
npm start
```
Access the application:
- Frontend: http://localhost:3000

🔧 Configuration

Backend Configuration

Edit src/main/resources/application.yml:

server:
  port: 8080
  servlet:
    context-path: /api

spring:
  datasource:
    url: jdbc:h2:mem:gympose_db  # Change to PostgreSQL for production
    username: sa
    password: password

jwt:
  secret: your-super-secret-jwt-key-change-in-production
  expiration: 86400000

Frontend Configuration

Edit src/services/api.js:

const API_BASE_URL = 'http://localhost:8080/api';

📱 Usage

Create an account with email and username
Secure JWT-based authentication
User profile management

2. Exercise Selection

Choose from available exercises:
- Knee Bend: Basic leg strengthening
- Squat: Full body compound movement
- Deadlift: Back and leg strength
- Lunge: Balance and unilateral training

3. Real-Time Pose Detection

Position your device camera
Start the exercise session
Get instant form feedback
Track repetitions and sets

4. Progress Monitoring

View workout history
Analyze performance trends
Track form improvement over time

🧠 How It Works

Pose Estimation Algorithm

Frame Capture: Real-time video stream from device camera
Landmark Detection: OpenCV-based pose landmark identification
Angle Calculation: Mathematical computation of joint angles
Form Analysis: AI-powered exercise technique evaluation
Feedback Generation: Instant coaching tips and corrections

Key Metrics

Knee Angle: Primary indicator for leg exercises
Hip Angle: Important for compound movements
Form Score: Overall technique quality (0-100%)
Confidence: Detection reliability score
Balance Score: Posture stability assessment

🗄️ Database Schema

Core Entities

Users: User accounts and profiles
ExerciseSessions: Workout session records
PoseData: Real-time pose analysis data
WorkoutPlans: Personalized exercise programs

Relationships

User (1) ←→ (Many) ExerciseSession
ExerciseSession (1) ←→ (Many) PoseData
User (1) ←→ (Many) WorkoutPlan

🔒 Security Features

JWT Authentication: Secure token-based authentication
Password Hashing: BCrypt encryption for user passwords
CORS Configuration: Cross-origin resource sharing setup
Input Validation: Comprehensive request validation
SQL Injection Protection: JPA/Hibernate security

📊 API Endpoints

Authentication

POST /api/auth/register - User registration
POST /api/auth/login - User login
GET /api/auth/profile - User profile

Pose Detection

POST /api/pose-detection/start-session - Start exercise session
POST /api/pose-detection/process-frame - Process video frame
POST /api/pose-detection/end-session - End exercise session
GET /api/pose-detection/history - Exercise history
GET /api/pose-detection/statistics - User statistics

Exercise Management

GET /api/pose-detection/exercise-types - Available exercises
GET /api/pose-detection/session/{id}/analysis - Session analysis

🎯 Exercise Types

1. Knee Bend

Target: Quadriceps, Hamstrings, Glutes
Form: Bend knees to 90° or less
Feedback: Depth and knee alignment

2. Squat

Target: Full body compound movement
Form: Lower body until thighs parallel to ground
Feedback: Depth, back position, knee tracking

3. Deadlift

Target: Lower back, Hamstrings, Glutes
Form: Hinge at hips, keep back straight
Feedback: Hip hinge, back position, bar path

4. Lunge

Target: Unilateral leg strength and balance
Form: Step forward, lower until both knees at 90°
Feedback: Balance, depth, knee alignment

🚀 Deployment

Production Backend

Build JAR file:
```
mvn clean package -DskipTests
```

Run with Java:

java -jar target/gympose-ai-backend-1.0.0.jar

Environment variables:

export SPRING_PROFILES_ACTIVE=production
export DATABASE_URL=jdbc:postgresql://localhost:5432/gympose
export JWT_SECRET=your-production-secret

Production Frontend

Build production bundle:
```
npm run build
```
Serve static files:
```
npx serve -s build -l 3000
```

🧪 Testing

Backend Tests

cd backend
mvn test

Frontend Tests

cd frontend
npm test

🤝 Contributing

Fork the repository
Create a feature branch
Make your changes
Add tests if applicable
Submit a pull request

📝 License

This project is licensed under the MIT License - see the LICENSE file for details.

🙏 Acknowledgments

OpenCV: Computer vision library
MediaPipe: Pose estimation framework
Spring Boot: Java application framework
React: Frontend framework
Tailwind CSS: Utility-first CSS framework

📞 Support

Issues: GitHub Issues
Documentation: Wiki
Email: support@gympose-ai.com

🔮 Future Enhancements

Mobile App: iOS and Android applications
Social Features: Share workouts and achievements
AI Coach: Personalized workout recommendations
Video Recording: Save and review workout sessions
Integration: Connect with fitness trackers and apps
Multi-User: Group workouts and competitions

Built with ❤️ for the fitness community

Transform your workouts with AI-powered coaching!

This site is open source. Improve this page.