It can be combined with various 3D rendering methods.It's very accurate because of the pixel precision of the canvas.

It stays sharp no matter how deep the zoom is and size of the device.

The images and texts are thus initially separated (before drawn) but as the page is shown, these objects are then blended together. Text and images are drawn on the canvas, in a resolution which suits the viewer or zoom level so that they always appear sharp.

HTML5 offers a number of ways of rendering text and images and while most web pages render images and HTML5 elements as objects, you can also use a canvas to do this which is more similar to what a painter would do if they were drawing a picture on a canvas. Open the page in a new tab and you'll see if text has been rendered together with images or not. If you're not afraid of HTML, then peek into the source code of the publication and you should see references to the image for each page. How do you ultimately know if this is what's going on? Try zooming deep into a page with your browser and check if the text stays sharp. Not only does it look bad, it also loads slowly. Text tends to start to blur eventually with images and text blending together in a bit of a pixelated mess (hello 1998!). You'll see if this is what they do if you try to zoom in deep enough into a text.

What does this mean? It means that each page in your PDF content will be converted into a JPEG or PNG image in high resolution in order to be shown on the web site. Some publishers claim to convert your content to HTML5 but they're not actually doing this, instead they're just bitmapping your pages.

The 'lets fake it until we make it' approach

How convert html to html5