Krezy - Cross-Platform Emotion Detection

A powerful and easy-to-use emotion detection library for Python that works seamlessly across web, mobile, and desktop applications.

✨ Features

• 🎯 Real-time Emotion Detection

  • Analyzes emotions from live video feeds
  • Supports webcam input
  • Fast and efficient processing

• 🔍 Multiple Detection Backends

  • OpenCV (default, fastest)
  • RetinaFace (most accurate)
  • MTCNN
  • SSD

• 📊 Rich Emotion Analysis

  • 7 basic emotions: Happy, Sad, Angry, Fear, Surprise, Disgust, Neutral
  • Percentage scores for each emotion
  • Dominant emotion detection
  • Timestamp tracking

• 🌐 Cross-Platform Support

  • Web applications (Flask)
  • Mobile apps (Kivy)
  • Desktop software (CustomTkinter)
  • Platform-specific optimizations

🚀 Installation

Install the package based on your needs:

# Core package (emotion detection only)
pip install krezy

# For web applications
pip install krezy[web]

# For mobile applications
pip install krezy[mobile]

# For desktop applications
pip install krezy[desktop]

# For all platforms
pip install krezy[all]

Note: Do NOT use pip install krezy_web - that's not the correct package name. The correct format is krezy[web].

📖 Quick Start

Basic Emotion Detection

from krezy import EmotionDetector

# Initialize detector
detector = EmotionDetector()

# Analyze an image
result = detector.detect_emotions('path/to/image.jpg')
print(result['emotions'])  # Shows emotion percentages
print(result['dominant_emotion'])  # Shows strongest emotion

Web Application

from krezy.web import create_app

# Create a Flask web app with emotion detection
app = create_app()
app.run(debug=True)

# Your emotion detection is now available at:
# - http://localhost:5000/ (web interface)
# - http://localhost:5000/analyze (API endpoint)

Mobile Application

from krezy.mobile import create_app

# Create a Kivy mobile app with emotion detection
app = create_app()
app.run()

Desktop Application

from krezy.desktop import create_app

# Create a modern desktop app with emotion detection
app = create_app()
app.run()

🎮 Advanced Usage

Custom Backend Selection

from krezy import EmotionDetector

# Use RetinaFace for better accuracy
detector = EmotionDetector(detector_backend='retinaface')

# Use MTCNN for balanced performance
detector = EmotionDetector(detector_backend='mtcnn')

Real-time Video Analysis

from krezy import EmotionDetector

detector = EmotionDetector()

# Start webcam analysis with visualization
detector.start_video_stream(camera_id=0, display_output=True)

📋 Requirements

• Python 3.7+
• OpenCV
• DeepFace
• Platform-specific requirements are handled automatically during installation

🔧 Troubleshooting

Common issues and solutions:

1. No face detected

   • Ensure good lighting conditions
   • Check if face is clearly visible
   • Try different detector backend

2. Performance issues

   • Use 'opencv' backend for speed
   • Reduce video resolution
   • Check system resources

3. Installation errors

   • Ensure Python 3.7+ is installed
   • Update pip: python -m pip install --upgrade pip
   • Install platform-specific dependencies first

🤝 Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

📄 License

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

🆘 Support

Need help? Try these resources:

• GitHub Issues
• Email: aakifmudel@gmail.com