Secure Gesture Project

Secure Gesture is an AI-powered system designed for face verification and hand gesture recognition. The project integrates webcam functionalities to authenticate users via facial recognition and enable control over laptop operations using hand gestures.

Project Structure

The project directory is organized as follows:

• face_verification: Includes modules and scripts responsible for face verification.
• hand_gesture: Contains modules and scripts related to hand gesture recognition.
• app.py: Main application file integrating face verification and hand gesture recognition.

Dependencies

• opencv
• numpy
• matplotlib
• tensorflow
• mtcnn
• keras-facenet
• sklearn
• pydot
• pyautogui
• mediapipe
• pandas
• seaborn

Usage

1. Face Verification:

   • Run collect_images.py to collect images using webcam used for training you model on you.
   • You should get a new folder called photos with nested folder named as your user_name submited when running the previous step.
   • Create a folder called Unkown inside photos folder and copy some photos from Labled Faces in the Wild dataset here, this will be used for training model on unkown persons. Note: don't copy all lfw dataset, just some random samples may be 300 is enough.
   • If you want you model to verify other persons beside your user_name you could run collect_images.py again with different user_names.
   • Follow instructions on main.ipynb notebook in order to train your model to classify your users by thier folder names inside photos dir.
   • After training your model you can run verification.py to see if the model can recognize and verify you.

2. Hand Gesture Control:

   • Once verified, perform hand gestures in front of the webcam to control laptop operations.
   • Supported gestures:
     • Adjust volume
     • Take screenshots
     • Open Text Editor
     • Painter