Skip to content

Facial landmark tracking and translation into 3D points with any regular video.

License

Notifications You must be signed in to change notification settings

ahzam1/CMPUT414Project

Repository files navigation

CMPUT414 Winter 2020

Authors: Ahzam Ahmad & Ziyuan Liu

To run the project:

  1. run extractLandmarks.py and rename the internal video source to whatever video you want.
  2. run interpolation.py
  3. open blender, and run the createMesh.py script from within.

singleImage.py is used for testing single images within the face and landmark detector. The haarMethod directory contains the work of using the Haar features, it is incomplete due to its poor performance, but is runnable. It currently uses the webcam, and you can see its performance in realtime.

About

Facial landmark tracking and translation into 3D points with any regular video.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages