Using HTML5 video with canvas

Using HTML5 video with canvas

You’ve already learned a lot about the <video> element in my previous article “What I’ve Learned From Working With HTML5 Video Over A Month”, but did you know that the <video> and <canvas> elements would be a great choice to use together? In fact, the two elements are absolutely wondrous when you combine them! I will show you a handy demo using these two elements, which I hope will prompt cool future projects from you. Firstly, place the <video> and <canvas>elements in your HTML. Give your dimensions(width, height) to <canvas> according to your video. And of course, hide your video with the hidden attribute or “display: none” styling. Because <canvas> will do the trick for you ;) Let’s go through them one by one.

View

<canvas width="1280" height="720"></canvas>
<video src="foo.mp4" preload="auto" autoplay playsinline webkit-playsinline muted hidden></video>
We’re hiding the video element because we just want <canvas> to show the video, not <video> element itself.

Declarations

Let’s declare the variables that will be used.
// These our <canvas> and <video> elements that will be used later.
const canvas = document.querySelector('canvas')
const video = document.querySelector('video')

// fps means frame per second. It’s important according to the video’s
// frame. You may want to use it as 24, or 30, or 60. It will be used
// for the intervals millisecond value.
const fps = 60

// width and height are the dimensions in pixels of the video. They
// will be given to the canvas element. If you do not specify them,
// <canvas> will take 300 for width and 150 for height as default
// which may break your view.
const width = 1280
const height = 720

// I will use canvasInterval variable to set and clear an interval to
// make <canvas> follow and unfollow the <video> element.  
const canvasInterval = null

Logic

Here is the magical part, with the whole code.
const canvas = document.querySelector('canvas');
const video = document.querySelector('video');
const fps = 60;
const width = 1280;
const height = 720;
const canvasInterval = null;

drawImage() {
  canvas.getContext('2d', {
    alpha: false
  }).drawImage(video, 0, 0, width, height);
}

canvasInterval = window.setInterval(() => {
  drawImage(video);
}, 1000 / fps);

video.onpause = function() {
  clearInterval(canvasInterval);
};

video.onended = function() {
  clearInterval(canvasInterval);
};

video.onplay = function() {
  clearInterval(canvasInterval);
  canvasInterval = window.setInterval(() => {
    drawImage(video);
  }, 1000 / fps);
};
drawImage() method draws the video onto the canvas. video.onpause(), video.onended(), and video.onplay() methods are related to the performance issue. When the video paused or ended, we do not need to draw <video> onto the<canvas>, because you know, the frame is not changing after paused or ended. And when the <video> starts playing/resuming, <canvas> should start/resume drawing the <video> onto <canvas> as well.