Skip to content
fanat503Public

About

Computer Vision project using MediaPipe, NumPy, CV2, . Gesture control & Face filters & Voice commands

Resources

Stars

1 star

Watchers

0 watching

Forks

Latest commit

 

History

15 Commits

Folders and files

Repository files navigation

Jarvis AI Assistant (v0.3)

This is a computer vision project powered by OpenCV and MediaPipe. The assistant can track faces, apply filters (nose mask) and take selfies using hand gestures.

Features

  • Face Tracking: Real-time face detection with a clown nose filter.
  • Hand Gestures: Show a "pinch" gesture (thumb + index finger) to take a selfie.
  • Multithreading: Smooth video capture without lag.

Requirements

  • Python 3.x
  • OpenCV
  • MediaPipe
  • NumPy

How to run

  1. Install libraries:
    pip install opencv-python mediapipe numpy
    
    

Controls 'n' - Toggle Nose filter 'b' - Toggle Face Blur 'q' - Quit

About

Computer Vision project using MediaPipe, NumPy, CV2, . Gesture control & Face filters & Voice commands

Resources

Stars

1 star

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages