let h = 480;

let video;

let poseNet;

let poses = [];

let skeletons = [];

​

function setup() {

  createCanvas(w, h);

  video = createCapture(VIDEO);

  // Create a new poseNet method with a single detection

  poseNet = ml5.poseNet(video, 'single', gotPoses);

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

  video.hide();

  fill(255, 0, 0);

  stroke(255, 0, 0);

}

​

function draw() {

  image(video, 0, 0, w, h);

​

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

  drawKeypoints();

  drawSkeleton();

}

​

// A function to draw ellipses over the detected keypoints

function drawKeypoints() {

  // Loop through all the poses detected

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

    // For each pose detected, loop through all the keypoints

    for(let j = 0; j < poses[i].pose.keypoints.length; j++) {

      // A keypoint is an object describing a body part (like rightArm or leftShoulder)

      let keypoint = poses[i].pose.keypoints[j];

      // Only draw an ellipse is the pose probability is bigger than 0.2

      if (keypoint.score > 0.2) {

        ellipse(keypoint.position.x, keypoint.position.y, 10, 10);

      }

    }

  }

}

​

// A function to draw the skeletons

function drawSkeleton() {

  // Loop through all the skeletons detected

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

    // For every skeleton, loop through all body connections

    for(let j = 0; j < poses[i].skeleton.length; j++) {

      let partA = poses[i].skeleton[j][0];

      let partB = poses[i].skeleton[j][1];

      line(partA.position.x, partA.position.y, partB.position.x, partB.position.y);

    }

  }

}

​

// The callback that gets called every time there's an update from the model

function gotPoses(results) {

  poses = results;

}