Hey guys, quick introduction. Visual Question Answering or VQA for short is an AI task. A system takes an image with a question as input, and predicts an answer to that question. An example could be a photo of a room and a question asking about location of an item such as wallet.
If this concept seems new, you can try a simple demo on my website
$1. First you upload an image, wait a few seconds and then ask questions. (The web seems to work fine with talkback, but since I don't have an IOS device, I would appreciate an Iphone user, that would try it and post a comment here, if it is usable with VoiceOver or not. Thanks in advance.)
Now back to the question. I'm certain many of you use smartphone apps which provide some kind of description of an image. That being said, are you using any app that provides VQA?