Hi all,
Is there a way to get the original video frame before it gets processed by the AR lib, modify it in some way and hand it back (maybe some callback func :confused: ).
Something like an preprocess stage before the AR takes control of the work?
I would like to add "Virtual :D Trackables" to the video stream.
All the best,
Alex.
Hi,
You are right regarding the need of virtual trackables. I just need to have the trackable registered once and then I have all that I need. KISS at its finest. :o
In the mean time a nice use case would be for testing purposes. record a movie and then just feed it to the AR lib. A nice black box testing for any app built upon this lib.
Another thing would be that you could apply any number of correction filters.
Anyway, I still think this stage in the pipeline should be accessible in future.
All the best,
Alex.
This isn't possible with the current SDK, the video goes directly to the tracker without any hooks for developer intervention. I'm trying to figure out the use case here... If you are able to draw virtual trackables into the scene, don't you already know their position and orientation? I suppose you might care about how they occlude other trackables, but again this is something you could figure out with a little math. If you know where all the "real" trackables are, and you know where you are drawing your virtual trackables, can't you achieve your goal?
- Kim
No way so far :(