I guess the reason we have QRs is because we are not yet there to determine what is the object in question. QR technology makes it easier by providing a unique tag for each object. So now when you see the QR code, your QR reader can access the look-up table to find what particular object it is.<p>The proposed technology is one step ahead of existing QR systems. But I guess the end goal to identify objects, and making a one to one association would be very cool. For Ex - when you want to identify people, you do not want them to come with QR code. Putting a name tag is something I will do to introduce myself in conferences so that people can recognize me from distance. But Ideally I would like that they recognize me by my face :)
The main programmer here.
For the technically curious, the ingredients of the image processing are OpenCV and CNNs(Convolutional neural nets).<p>On design aspects, we wrote a short post here: <a href="http://blog.taglev.com/design-philosophy-behind-taglev/" rel="nofollow">http://blog.taglev.com/design-philosophy-behind-taglev/</a>
QR has a lot of error correction built in which this solution doesn't offer. Also, you would need a central repository to hold all the tags as they are just references and not the final URLs themselves.
But that might be OK considering all short URL services work the same way.
The "hitler finds out about X" meme isn't played out yet? I was quite surprised to see that. Also I find the quote at the bottom in poor taste, especially with the exclamation point. But to each their own.