Log in or register to post comments

3D recognition 2016

August 7, 2016 - 6:44am #1

Hi Vuforia team,

Can you please share some updates on the evolution of Vuforia 3D recognition. 

We are lookingto achieve this:
https://www.youtube.com/watch?v=BipmUstUMjk

Is Vuforia capable of delivering this applications?

Did you had any success on using the object scanning sheet of large objects such as a car? 

Did the technology evolved in this sense since july 2015?

Thanks,

3D recognition 2016

August 8, 2016 - 10:28am #2

Hello,

The capabilities of the Object Recognition feature remains largely similar to when it was first released. In the video, I observed two, distinct CV features being demonstrated:

  1. Markerless AR - The parts of the video when the tire went into an 'exploded' view
  2. Large-scale object recognition - The parts of the video when the existing car was augmented

For #1, Vuforia is actively evaluating this feature and hopes to be able to commercialize a solution for this use case in the near future.

For #2, although scanning a full-sized car with the Vuforia Object Scanner is technically feasible, there are limitations in the scanning process that make it challenging:

https://developer.vuforia.com/forum/object-recognition/detecting-large-objects

My response to the post above should help to highlight these.

Have you considered Vuforia Studio for #2?

https://developer.vuforia.com/studio.

I understand that it still uses markers (unlike the video you've provided), but can integrate CAD data to help with authoring and augmentations.

Thanks.

Log in or register to post comments