Slides: https://docs.google.com/presentation/d/17_Vxjl7UcR18hSdGtug01ekwg4kwNhlOHB1PghZuBtw/edit?usp=sharing
Teachable machine Part selections with the keywords: https://editor.p5js.org/ml5/sketches/PoseNet_part_selection
https://editor.p5js.org/ml5/sketches/PoseNet_webcam (BASIC)
https://editor.p5js.org/ambika/sketches/nSyIG-UCO (text with two wrists)
https://editor.p5js.org/ambika/sketches/oo9-GesSH (flip image)
https://editor.p5js.org/ambika/sketches/C2wUeZQZ_ (two poses and noses)
Ml5 data provenance: https://github.com/ellennickles/ml5js-model-and-data-provenance-project
Dan Oved's post: https://blog.tensorflow.org/2018/05/real-time-human-pose-estimation-in.html