Facial Verification App 📷 🧐

This README file provides a detailed explanation of the Kivy facial verification app, outlining its functionalities, usage instructions, and dependencies .

Project Overview 🎨

This application leverages Kivy, OpenCV, and DeepFace to create a real-time facial verification system using your webcam . It captures frames from the camera, detects faces using DeepFace's superpowers , and compares them against a predefined database (not included in this code). If a match is found, the user's identity is displayed which can be processed to do further functions. Otherwise, a red message indicates a no match or face detection failure.

Features 🚀

• Real-time webcam capture
• Face detection using DeepFace's magic ✨
• Facial verification against a user database (needs database of image for each user inside the train repo, )
• Visual feedback with success/failure messages and color changes (green for success, red for errors )

Dependencies

• Kivy: Cross-platform GUI framework (https://kivy.org/)
• OpenCV: Real-time computer vision library (https://opencv.org/)
• DeepFace: Deep learning-based facial recognition library (https://github.com/serengil/deepface)
• pandas : Data manipulation library (https://pandas.pydata.org/)

Installation

1. Ensure you have Python (version 3.x recommended) installed on your system (check with python --version in your terminal).

2. Create a virtual environment to manage dependencies (optional but recommended, keeps things tidy! ).

3. Install the required libraries using pip:

   pip install kivy opencv-python deepface

4. (Optional) Install pandas and numpy if you intend to modify the face detection functionality.

Usage 🖥️

1. Create a 'Train' repository in the main directory and include a class of users which contain image of the user. do this for each users.(not included in this code, but you can set it up later! )

2. (Optional) resize the iamges inside each repo to the same shape. This will help the model learn on a uniform data.

   python img_resizer.py

3. Run the application from your terminal:

   python main.py

Code Structure

The code is organized into the following classes and functions:

• CamApp class:

  • Initializes the application, including webcam capture and layout elements (creates the user interface ).
  • Provides methods for updating the webcam image, face detection, verification, and label reset (keeps things running smoothly! ).

• face_detect function:

  • Uses DeepFace to detect faces in an image and compare them against the database (needs implementation, but the code is ready to accept your custom logic! ).
  • Returns the identified user's information or a message indicating no match or face detection failure.

• verify function:

  • Captures a frame from the webcam, saves it as an image, performs face detection, and updates the verification label based on the outcome (the verification magic happens here! ✨).

• reset_verification_label function:

  • Resets the verification label to its default state (green text, "Welcome") after a verification attempt (clears the message after verification is done! ).

Additional Notes

• The code assumes you have a pre-defined user database set up for verification.
• You might need to adjust the DeepFace model selection (models[0]) based on your requirements and installation.
• Consider error handling and logging for robustness (makes the code more reliable ).

Contributing 💡

Contributions are welcome! If you have any ideas for improvements or new features, feel free to open an issue or submit a pull request.

Credits 🙌

This project was created by Dawit Neri

Support 💬

If you encounter any issues or have any questions, feel free to reach out to dawitneri888@gmail.com or open an issue in the GitHub repository.