Yuva is an innovative mobile application designed to identify and nurture athletic talent across India, especially in rural and remote areas. Powered by on-device AI, Yuva enables standardized athletic testing without the need for specialized equipment or expert scouts.
WhatsApp.Video.2025-09-15.at.2.03.34.PM.mp4
WhatsApp.Video.2025-09-17.at.8.53.19.AM.mp4
Identifying athletic talent in a vast country like India is a monumental challenge. Aspiring athletes in rural and remote areas lack access to standardized testing facilities and expert scouts. This leads to a massive loss of potential talent that could otherwise excel in various sports disciplines.
- 📱 On-Device AI – Analyzes performance videos offline directly on the user's phone
- 🏃 Test Battery – Guides users through SAI's standard tests (vertical jump, sprint, sit-ups, etc.)
- 🔍 Cheat Detection – Flags tampered videos and incorrect form using advanced algorithms
- 📊 Benchmarking – Instant comparison with national standards and age-group percentiles
- 🎮 Gamification – Badges, leaderboards, and progress tracking to maintain engagement
- 🔒 Secure Submission – Encrypted upload to SAI servers when connectivity is available
- 🌐 Multi-Language Support – Accessibility in various Indian languages
Yuva implements the Sports Authority of India (SAI) standard tests:
-
Physical Assessments
- Vertical Jump (explosive leg power)
- Sprint Test (acceleration and speed)
- Agility Test (quickness and coordination)
- Endurance Run (cardiovascular fitness)
-
Strength Tests
- Push-ups (upper body strength)
- Sit-ups (core strength)
- Pull-ups (back and arm strength)
- Plank Test (core endurance)
-
Skill Assessments
- Ball Control (sport-specific dribbling skills)
- Shooting Accuracy (precision and power)
- Reaction Time (quick response ability)
- Coordination (body movement harmony)
- Frontend: React Native with Expo for cross-platform compatibility
- Styling: React Native StyleSheet with custom components
- AI Model: TensorFlow.js with MoveNet for pose estimation
- Animations: React Native Reanimated for smooth UI interactions
- Deployment: Mobile app stores (iOS & Android) with web-based admin panel
- Video Recording: The app guides the user to record a specific fitness test with clear instructions
- On-Device Processing: TensorFlow Lite models analyze the video frame-by-frame on the user's phone to:
- Detect human pose and key body points
- Calculate performance metrics (jump height, sprint time, rep count)
- Run cheat detection algorithms to ensure test validity
- Instant Feedback: Results are calculated, benchmarked, and displayed to the user immediately
- Secure Sync: Verified and encrypted data packets are synced with the SAI backend when connectivity is available
- Clone the repository:
git clone https://github.com/your-username/yuva-talent-scout.git
cd yuva-talent-scout- Install dependencies:
npm install
# or
yarn install- Install TensorFlow.js dependencies:
expo install @tensorflow/tfjs @tensorflow-models/posenet- Start the development server:
expo start
# or
npm start- Run on your preferred platform:
- Press
afor Android emulator - Press
ifor iOS simulator - Scan QR code with Expo Go app on your physical device
- Registration: Create an account with basic details and age verification
- Test Selection: Choose from the standardized test battery
- Guidance: Follow step-by-step instructions for each test
- Recording: Capture performance using phone's camera with on-screen guidance
- Analysis: Get immediate AI-powered feedback and results
- Tracking: Monitor progress over time and compare with national standards
yuva-talent-scout/
├── assets/ # Images, fonts, and model files
├── components/ # Reusable UI components
│ ├── CameraModule/ # Custom camera with guidance overlay
│ ├── PoseRenderer/ # AI pose estimation visualization
│ ├── NavBar/ # Bottom navigation component
│ └── ProgressChart/ # Data visualization components
├── screens/ # Main app screens
│ ├── HomeScreen.js # Dashboard with stats and progress
│ ├── TestScreen.js # Athletic assessment tests
│ ├── TutorialsScreen.js # Test instructions and demonstrations
│ ├── ProfileScreen.js # User profile and historical data
│ └── LeaderboardScreen.js # Gamification and rankings
├── services/ # Business logic and external services
│ ├── ai-analysis/ # TensorFlow.js integration
│ ├── encryption/ # Data security utilities
│ └── api/ # SAI server communication
├── utils/ # Helper functions and utilities
│ ├── cheat-detection/ # Algorithm for identifying tampered tests
│ └── benchmarks/ # National standard comparisons
└── App.js # Main application component
To adapt Yuva for different regions or sports:
- Modify Test Parameters: Adjust the
testConfigurationsfor different age groups or regions - Add New Tests: Extend the test battery with sport-specific assessments
- Update Benchmark Data: Modify national standards in the benchmarks utility
- Regional Adaptation: Add multi-language support for different regions
We welcome contributions to Yuva! Please feel free to submit pull requests, report bugs, or suggest new features.
- Fork the project
- Create your feature branch (
git checkout -b feature/AmazingFeature) - Commit your changes (
git commit -m 'Add some AmazingFeature') - Push to the branch (
git push origin feature/AmazingFeature) - Open a Pull Request
- Follow React Native best practices
- Ensure accessibility standards are met
- Maintain test coverage for critical functionality
- Document new features thoroughly
- Sports Authority of India (SAI) for test protocols and standards
- TensorFlow.js team for pose estimation models
- React Native community for excellent documentation and support
- The many coaches and athletes who provided feedback during development
