let video;

let hands = [];

​

function setup() {

  createCanvas(640, 480);

  video = createCapture(VIDEO);

  video.size(width, height);

​

  handpose = ml5.handpose(video, modelReady);

​

  // This sets up an event that fills the global variable "predictions"

  // with an array every time new hand poses are detected

  handpose.on("hand", results => {

    hands = results;

  });

​

  // Hide the video element, and just show the canvas

  video.hide();

}

​

function modelReady() {

  console.log("Model ready!");

}

​

function draw() {

  image(video, 0, 0, width, height);

​

  // We can call both functions to draw all keypoints and the skeletons

  drawKeypoints();

}

​

// A function to draw ellipses over the detected keypoints

function drawKeypoints() {

  for (let i = 0; i < hands.length; i += 1) {

    const hand = hands[i];

    for (let j = 0; j < hand.landmarks.length; j += 1) {

      console.log(hand)

      const keypoint = hand.landmarks[j];

      fill(0, 255, 0);

      noStroke();

      ellipse(keypoint[0], keypoint[1], 10, 10);

    }

  }

}