Log in or register to post comments

Selection algorithm in case of several targets

January 15, 2014 - 2:05am #1

Hi,

I would like to get some information regarding how things work when there are more than one target in front of the camera at the same time.

Specifically:

1. How does Vuforia determine which target is identified first? Can I assume, for example, that the first trackable result object in the array (that is available via the state object)  is the closest to the camera or something like that?

2. In case I put the camera in front of two targets (at the same time), will the detection be done at the same time? (meaning - will I have the detection of both at the same rendering frame?) If not, can I assume something regarding the time it takes to detect both?

 

Thanks,

yakobom

Selection algorithm in case of several targets

January 15, 2014 - 10:47am #5

Well, basically I want to respond only to one of the several targets that are in front of me, based on specific parameters. So I undertsand I will have to write my selection logic for selecting the one I want, but for that I need to know that I detect them at the same time and that.

I will find the way to do this.

Thank you for your kind assistance.

 

yakobom

Selection algorithm in case of several targets

January 15, 2014 - 10:42am #4

The sequence of detection is effectively not predictable, the factors you describe are not determinative, though you can expect that a target with a higher rating will be detected before one with a lower rating - all other factors being equal. 

What is it that you are trying to accomplish?

Selection algorithm in case of several targets

January 15, 2014 - 10:02am #3

Thanks for your reply.

That's a bit disappointing...

Is there any guidline at all that you can supply on this regards? Or is it simply not predictable?

 

yakobom

Selection algorithm in case of several targets

January 15, 2014 - 2:18am #2

1. How does Vuforia determine which target is identified first? Can I assume, for example, that the first trackable result object in the array (that is available via the state object)  is the closest to the camera or something like that?

No, you cannot make any such assumption; there is no general rule to foresee which target will be dtected first (assuming they are both simultaneously visible in the camera view)

2. In case I put the camera in front of two targets (at the same time), will the detection be done at the same time? (meaning - will I have the detection of both at the same rendering frame?) If not, can I assume something regarding the time it takes to detect both?

No, again, you cannot predict or make assumptions on this.

 

 

Log in or register to post comments