Skip to content

ZBZFirst/HandFaceTracker

Repository files navigation

Gesture-Controlled Face & Hand Tracking (Python)

This project uses MediaPipe and OpenCV to perform real-time face and hand tracking from a webcam.
Detected gestures can optionally control OBS Studio via WebSocket.


Features

  • Real-time face landmark detection
  • Real-time hand gesture recognition
  • Visual overlay of landmarks
  • Optional OBS Studio control (filters, scene effects)
  • Graceful handling when optional components are missing

System Requirements

Operating System

  • Windows 10/11 (recommended)
  • macOS / Linux may work with camera backend changes

Python

  • Python 3.9 – 3.11
  • Python 3.10 is strongly recommended

Check version:

python --version


opencv-python>=4.8,<5
numpy>=1.23,<2.0
mediapipe>=0.10.9
obs-websocket-py>=1.0
#i think?

About

A Hand and face tracker for Python

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages