🎭 DanceMoveInsight

A Flask-based REST API for analyzing dance poses in videos using MediaPipe and OpenCV.

✨ Features

• 🎥 Video Upload: Support for multiple video formats (MP4, AVI, MOV, etc.)
• 🕺 Pose Detection: Real-time dance pose analysis using MediaPipe
• 🔌 REST API: Clean JSON API endpoints
• 🌐 Web Interface: Simple HTML upload interface
• 🐳 Docker Support: Containerized deployment ready
• 💾 Result Storage: Persistent storage of analysis results

🚀 Quick Start

🌐 Live Demo (Try it now!)

🎭 Test the API live at: http://51.21.190.200:80

• Upload a dance video and see the pose analysis in action
• No setup required - just visit the URL and start testing!

Using Docker (Recommended)

# Run the pre-built image
docker run -p 5000:5000 dancemoves:latest

# Or build from source
docker build -t dancemoves .
docker run -p 5000:5000 dancemoves

Local Development

# Install dependencies
pip install -r requirements_flask.txt

# Run the Flask app
python flask_api.py

Access the API at http://localhost:5000

📁 Project Structure

DanceMoveInsight/
├── 📄 Core Application Files
│   ├── flask_api.py              # Main Flask API server
│   ├── dance_pose_detector.py    # Pose detection logic
│   ├── config.py                 # Configuration settings
│   └── requirements_flask.txt    # Python dependencies
│
├── 🐳 Docker & Deployment
│   ├── Dockerfile                # Docker container configuration
│   └── deployment/               # Deployment scripts and configs
│       ├── aws/                  # AWS ECS deployment
│       ├── scripts/              # Deployment automation
│       └── docker-compose.yml    # Local development setup
│
├── 📚 Documentation
│   └── docs/                     # All documentation files
│       ├── API_README.md         # API documentation
│       ├── DEPLOYMENT.md         # Deployment guide
│       └── FLASK_API_SUMMARY.md  # Technical summary
│
├── 🧪 Testing
│   └── tests/                    # Test files
│       ├── test_api.py           # API endpoint tests
│       └── test_dance_pose_detector.py  # Core logic tests
│
├── 💡 Examples
│   └── examples/                 # Example usage and demos
│       ├── example_usage.py      # Usage examples
│       └── demo_json_output.py   # Output format demos
│
├── 🎨 Frontend
│   └── templates/                # HTML templates
│       └── upload.html           # Web upload interface
│
└── 📊 Results
    └── results/                  # Analysis results storage

🔌 API Endpoints

Method	Endpoint	Description	Live Demo
POST	/upload	Upload video for pose analysis	Try it
GET	/health	API health check	Check Status
GET	/	Web upload interface	Web UI

🕺 Detected Poses

1. Arms Up: Both arms raised above shoulders
2. Arms Crossed: Arms crossed in front of body
3. One Arm Up: Single arm raised while other is down
4. Squat: Knees bent in squatting position
5. Lunge: One leg forward with bent knee, other leg back
6. Arabesque: One leg raised behind the body
7. Standing: Basic upright standing position

📋 Supported Formats

• Video: MP4, AVI, MOV, WMV, MKV, FLV, WEBM
• Max Size: 100MB
• Recommended: MP4 for best compatibility

🛠️ Development

# Run tests
python -m pytest tests/

# Start development server
python flask_api.py

# Run examples
python examples/example_usage.py

🧪 Test the Live API

# Test health endpoint
curl http://51.21.190.200:80/health

# Upload a video for analysis
curl -X POST -F "video=@your_dance_video.mp4" http://51.21.190.200:80/upload

📖 Documentation

• API Guide: docs/API_README.md
• Deployment: docs/DEPLOYMENT.md
• Examples: examples/

🔧 Requirements

• Python 3.8+
• OpenCV 4.8+
• MediaPipe 0.10+
• Flask 3.0+

---

🎬 Try It Now!

🌟 Live Demo: http://51.21.190.200:80

Ready to analyze some dance moves? 🕺💃

1. Quick Test: Visit the live demo above
2. Local Setup: docker run -p 5000:5000 dancemoves:latest
3. Development: Clone this repo and start coding!

📱 How to Use the Live Demo:

1. Visit http://51.21.190.200:80
2. Upload a dance video (MP4, AVI, MOV, etc.)
3. Wait for the analysis to complete
4. View the detailed pose detection results!

Example videos to try:

• Short dance clips (10-30 seconds work best)
• Clear view of the dancer
• Good lighting conditions