Skip to content

cyb3rkun/MPVMC

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

6 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Features:

  • 51 ARKit Blendshapes: Perfect Sync facial tracking with a webcam
  • Head Transform: Translation and Rotation with Quaternions.
  • Works on Linux

Setup

  1. Clone the repo:

    git clone https://cyb3rkun/MPVMC.git
  2. Setup Virtual Environment:

    cd MPVMC
    python3 -m venv MPVMCenv
    source ./MPVMCenv/bin/activate
    pip install mediapipe numpy python-osc python opencv-python
  3. Run the script:

    python facetracker.py
  4. configure you're software:

    • IP: 127.0.0.1
    • PORT: 39539
  5. Face Landmarker Task:

    • This repo contains the mediapipe task file v2 that supports blendshapes and head transform matrixes. no need to download it manually,
    • If you want to download it manually, you can do so whith this command:
       cd MPVMC && \
       wget -O ./assets/face_landmarker_v2_with_blendshapes.task -q https://storage.googleapis.com/mediapipe-models/face_landmarker/face_landmarker/float16/1/face_landmarker.task
    • for more information visit: Mediapipe Python Example Code for Face Landmarker or collab.research
  6. Enjoy you're vtubing journey with mediapipe on linux :)

About

A simple python script for mediapipe face tracking over VMC. it was created for warudo, but should work with any application that can recieve VMC input

Resources

License

Stars

0 stars

Watchers

0 watching

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages