Video from Review Session November 10th, 2022

In this session we cover:

• Review on how to examine variables in the console
• Discussion on PoseNet and AI trained body tracking
• Javascript Callback Functions
• Capturing Video
• Using PoseNet and ml5 to identify body parts and the 3 ways data is returned:
  • Points
  • Skeleton
  • Body part properties
• Creating the basics of a game that is controlled with your hand

Code2BodyTracking.mp4

Examining Variables and Objects in the Console

The PoseNet module returns data that is a little more complicated than what we have been used to in the past. As a review, you should play with data and see what it looks like when we log it to the Debug console in the P5 Editor

Code from our review session:

https://editor.p5js.org/des8963/sketches/ovDxklBgR

PoseNet, how does it work?

Behind the scenes PoseNet uses an Artificial Intelligence system called TensorFlow which has been trained on a large number of images of bodies with known positions of body parts. For more information on this, take a look at the following: