🏋️‍♂️ Fitverse - AI-Powered Fitness & Health Platform

Fitverse is an AI-driven web platform delivering personalized fitness and wellness solutions through computer vision and machine learning. The system provides real-time form correction, adaptive workout/diet planning, and specialized women's health features.

Example: Real-time pose detection interface

✨ Core Features

AI-Powered Modules

Feature	Technology	Description
Real-Time Pose Detection	TensorFlow.js + MediaPipe	Live exercise form analysis with WebGPU acceleration
Personalized Workouts	ML Recommendation Engine	Adaptive exercise plans based on goals/progress
Nutrition Planning	NLP + Dietary APIs	Dynamic meal plans with calorie/macro tracking
Injury Prevention	Biomechanics Analysis	Form correction alerts and alternative exercises
Women's Health	Cycle Prediction Algorithms	Period tracking + pregnancy-safe exercise regimens

Integrated Services

• BMI Calculator - Body composition analysis
• AI Nutritionist - Supplement/food recommendations
• Expert Consultations - WebRTC video coaching (PeerJS)
• Health Dashboard - Progress visualization (Recharts)
• Mobile-First Design - Responsive TailwindCSS layouts

System Architecture

graph TB
    subgraph Frontend
        A[React 19 UI] --> B[State Management];
        B --> C[Real-Time Visualization];
        C --> D[WebGPU Acceleration];
    end
    
    subgraph AI Engine
        E[TensorFlow.js] --> F[Pose Detection];
        E --> G[Recommendation Engine];
        F --> H[MediaPipe Integration];
        G --> I[ML Algorithms];
    end
    
    subgraph Backend Services
        J[Node.js API] --> K[PostgreSQL];
        J --> L[Redis Cache];
        J --> M[AWS S3 Storage];
        J --> N[WebRTC Signaling];
    end
    
    A --> E;
    A --> J;
    F --> D;

Loading

Technical Stack

Frontend

• React 19 (Concurrent Rendering)
• TailwindCSS 4 + CSS Modules
• State Management: React Context API
• Visualization: Recharts
• Animations: Framer Motion

AI/Computer Vision

• TensorFlow.js v4.9.0
• @tensorflow-models/pose-detection
• MediaPipe Pose Solution
• TFJS WebGPU Backend

Communication

• REST API: Axios
• Real-Time: PeerJS (WebRTC)
• WS Protocol: Socket.IO

Backend

• Node.js + Express
• PostgreSQL (User Data)
• Redis (Session Cache)
• AWS S3 (Media Storage)

📂 Project Structure (Expanded)

src/
├── ai/
│   ├── pose-detection/       # TF.js pose estimation logic
│   ├── recommendation/       # ML model handlers
│   └── utils/                # Tensor processing helpers
├── api/                      # Axios API clients
├── assets/                   # Media resources
├── components/
│   ├── exercise/
│   │   ├── detector/        # Pose detection components
│   │   ├── DeskExercise/    # Workspace-friendly workouts
│   │   ├── ExerciseCard.jsx # Workout UI cards
│   │   └── form-correction/ # Biomechanics feedback
│   ├── health/
│   │   ├── PeriodTracker.jsx # Cycle prediction
│   │   └── PregnancyCare.jsx # Prenatal guidance
│   └── dashboard/           # Metrics visualization
├── contexts/                # React context providers
├── hooks/                   # Custom React hooks
├── layouts/                 # Page templates
├── services/                # Business logic
├── utils/                   # Helpers & configs
└── views/                   # Page components

Screenshots

Desktop Views

Mobile Responsive Views

Installation & Development

Prerequisites

• Node.js v18+
• Yarn v1.22+
• TFJS-compatible GPU (recommended)

# Clone repository
git clone https://github.com/Sameer01-01/inhouseproject_final.git
cd inhouseproject_final

# Install dependencies
yarn install

# Configure environment
cp .env.example .env.local
# Update API keys in .env.local

# Start development server
yarn dev

# Build for production
yarn build

Deployment

Production build optimized with Vite:

yarn build && yarn preview

Docker deployment:

FROM node:18-alpine
WORKDIR /app
COPY package.json .
RUN yarn install --frozen-lockfile
COPY . .
RUN yarn build
EXPOSE 5173
CMD ["yarn", "preview"]

Performance Metrics

Module	Loading Time	Model Size	FPS
Pose Detection	1.8s	8.4MB	42fps
Diet Planner	0.6s	2.1MB	N/A
Period Tracker	0.3s	0.4MB	N/A

License

MIT License - See LICENSE.md for details.
External model assets (MediaPipe) subject to Google's Terms

Future Roadmap

1. Mobile Application - React Native port (Q4 2025)
2. Wearable Integration - Apple Watch/Google Fit sync
3. Advanced Biomechanics - 3D motion capture analysis
4. Voice Coaching - Real-time audio feedback
5. Multi-language Support - i18n implementation
6. Health API - Apple Health/Google Health Connect

timeline
    title Fitverse Development Timeline
    section 2025
        Q3 : React Native Mobile App
        Q4 : Wearable Integration
    section 2026
        Q1 : 3D Motion Capture
        Q2 : Voice Coaching
        Q3 : Multi-language Support
        Q4 : Health API Ecosystem

Loading

---

▶️ Watch Product Demo
📸 View Screenshot Gallery