An intro to object detection on MediaPiPe and the inner workings of Project GameFace

Опубликовано 13 мая 2024, 22:42

Project GameFace is an open-source hands-free gaming mouse powered by AI that allows users to control a computer cursor using their head movement and facial gestures. Discover how MediaPipe is used in Project GameFace to capture movements with a webcam and is recognized by ML models to move the cursor accordingly. Paul Ruiz, a Developer Advocate at Google, and Lance Carr, a Gamer and the inspiration behind Project GameFace, talk about object detection, another tool in the MediaPipe toolkit

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

Chapters:
0:00 Intro
0:56 How stickers helped development
1:38 Data collecting
4:07 Utilizing the data pool
6:09 Using MediaPipe studio
7:58 Running Object Detection
9:07 What applications can use this tech?
9:48 Practical applications
10:53 More info about GameFace

Subscribe to the TensorFlow channel → goo.gle/TensorFlow

#MediaPipe #Accessibility #Gaming


Speaker: Paul Ruiz