It'll be so useful to see contextual advertising near all of my favorite shops with the great deals that I love -- all without having to clip coupons! It'll be great when this is paired with facial recognition, so that I can see just the ads that I need for my personal situation at that moment. Maybe they'll even give me a better deal since I'm such a frequent customer!
Two thumbs up!
I just wait for someone to fall under the car collecting gold circles...
EDIT: Don't get me wrong, AR have a huge potential but not for flying emojis or playing RPG games running all over the city like a maniac with a huge phone.
Compared to what? This sounds like something a closed minded person would say. My grandmother says "I cant believe you stare at screens all day". Today, people my age don't even question it.
The phones years ago were too old, and the software was pretty unusable, but now the hardware is significantly better, you can push a 100,000 poly model in realtime over 30 fps video and it's smooth. Have you used the BBC app that allows you to walk around the mummy tomb? It's smooth and gorgeous and really fun.
I'd have to say that their promo video was pretty uninspiring, I'd like to have seen a spatially aware map-type demo, or some higher poly models, or a simple RPG game. The floating 2D emoji sprites don't do the potential justice, but hey, they're just starting out!
For instance compared to talking with each other. Didn't you spot that at every bar? Snapchat or Facebook Messenger are the main sources of communication now even for people who sit side by side. But probably I'm closed minded and I do not see a bigger picture.
They have gone round and taken a video of every street in a certain area, unpacked it, extracted salient points, reconstruct those points to get a 3d map.
From that, given any 2d image you should be able to extract a bunch of "salient points" or known points, which from their relationship to each other can tell where the camera is, and what direction its pointing.
The two hard parts are 1) collecting the data 2) searching the data in reasonable amount of time
https://youtu.be/tXwVg2S9wuY?t=60
the "salient points" are called keypoints and their feature vectors are called descriptors
https://en.wikipedia.org/wiki/Scale-invariant_feature_transf...
you are correct that the challenge is collecting and indexing/retrieving but there have been techniques that do this for a while
https://web.eecs.umich.edu/~michjc/papers/p144-park.pdf
(they even tested against SIFT descriptors)
the real thing that i'm puzzled by with blue vision is how they're registering against ARKit descriptors (if they are at all) since apple doesn't expose them in the ARKit api (only the point cloud itself). ARCore used to expose them (https://stackoverflow.com/a/29012790) but i don't think it does anymore. they must be doing the registration because they only support devices that are running ARKit/ARCore (and without it they would just have built a SLAM system - albeit backed by an "arcloud" - that sits beside ARKit/ARCore and would most likely be inferior).
The SDK will initially be free to use"initially"
/s