Write For Us

An introduction to MediaPipe’s Face Landmarker & how it can power hands free gaming

E-Commerce Solutions SEO Solutions Marketing Solutions
40 Views

  • Thanks! Share it with your friends!

    You disliked this video. Thanks for the feedback!

ava
Published
Paul and Lance dive deep into Face Landmarker, the underlying MediaPipe task that's used for Project Gameface and dissect the code piece by piece. Paul explains how the model works to interpret extraneous movements that you can make with your face and how it protects user privacy. Learn how you can use the model to power your own creations.

Project Gameface is an open-source hands-free gaming project powered by AI that allows users to control a computer using their head movement and facial gestures. For more info, check out the first episode.

Resources:
Learn more about Project Gameface → https://goo.gle/3wCCaDR
Watch episode 1 → https://goo.gle/3WGc5yd
Watch episode 2 → https://goo.gle/4b77mtX
Watch episode 3 → https://goo.gle/3UXswVA

Chapters:
0:00 - Introduction
0:50 - Last episode recap
2:07 - What is tessellation
4:14 - Face landmarkers
6:46 - Creating the detector
7:33 - Detection results
8:37 - What is the lag?
9:01 - Highlighting the irises
9:26 - How the model is open to any face
10:38 - Open for anyone to use

Subscribe to the TensorFlow channel → https://goo.gle/TensorFlow

#MediaPipe #Accessibility #Gaming

Speaker: Paul Ruiz
Category
Project
Tags
Google, developers, pr_pr: TensorFlow;
Sign in or sign up to post comments.
Be the first to comment

Up Next