Because it's not always pages and a modern smartphone camera can easily compete with a consumer-grade flatbed-scanner. I actually do own one of those, with a feeder (though not duplex) and it's a hassle compared to using Scannable. My current setup - excuse the mess on my desk, please - just required an additional $8 adapter for an run-of-the-mill tripod, but other smartphone holders, like a gooseneck clamp work too. Put Scannable on Default capture and start putting your items on the desk - papers, letters, book covers, etc - and Scannable will snap them reliably. But it's unforgiving when there is a small distraction - your attention comes back and you find dozens of identical pictures in your session.
Hi… Default capture usually works fine, but it's a bit of a bother when you use a tripod or anything else that keeps your smartphone stationary. Use case: Scanning lots of papers, covers, whatever. Scannable discovers the edges, take the picture and… proceeds to take the next photo. Of the same thing. Unless you pay attention and yank it away at once it had been processed. It would be better if Scannable would take the photo and then wait until it sees something new. I think this could be done entirely with the code you have now, as any picture is defined by the four angles of the rectangle scannable detects, And the lengths of their sides. As long as the phone is stationary, those won't change until the user takes the paper anyway and places another one.