Start a new topic
Solved

Multiple Images Recognition Simultaneously

Hi Team,

I'd like to know if it is possible to detect multiple targets at the same time. By this I mean overlay content on two or more different targets simultaneously.

With the sample apps (JS iOS & Native iOS), I'm not able to overlay content on both the surfer AND the biker images at the same time.

Is it possible to have it so that content is overlaid on both images at the same time, by perhaps instantiating multiple Client Trackers within a class?


See attached image. It shows only the surfer image being interacted with.

IMG1112.PNG
(1.87 MB)

Hello Andy,


Unfortunately tracking and augmenting multiple image targets simultaneously is currently not supported.

Can you describe in more detail what you are trying to acheive? There may be alternative ways to acheive your goals.


Thanks

Eva

Hi Eva, thank you for replying.


I'm attempting to create an app for an art gallery, where the art pieces are relatively close together. 

I'd like to provide rich content overlaid on these items (video etc.), or buttons that could point to URL links and resolve in a web view on the user's device.

I can't think of other potential solutions apart from multiple image target augmentation, but if you have further ideas, please let me know.

Thanks,
Andy.

Hello Andy,


What you could do is create additional target images for those art pieces that are placed close to each other. So for example, you have an image target for Art Piece#1 and an image target for Art Piece#2. If these two pieces are placed really close to each other you would want to recognise each piece at the same time and place relative information. In order to do that, you could create a third image collection that would contain Art Piece#1 and Art Piece#2 and place relative information for each art piece.


Thanks

Eva

Hello,

Great news for you as today we launched SDK 7, which introduces as well your feature request, “Multiple Image Target”. In a nutshell, SDK 7 includes:

  • Object Recognition
  • Multiple Image Target Recognition
  • Hit-testing API for SLAM
  • Instant tracking improvements
  • Extended recognition range

Multiple Image Target’ feature enables recognition of several images simultaneously. Once the images are recognized, developers will be able to layer 3D models, buttons, videos, images and more on each target. Additionally, augmentations will be able to interact with each other based on the targets’ positions.


Check out this blog for a developer insight. You can download the new SDK 7 directly from our Download page. Enjoy our new cool features ;)

Thanks
Eva

 

Hi,


Does this new feature work if the two images are the same ? For example, if I have two cards "Diplodocus" in my field of view, will I see two diplodocus ?

Hello Carl,

With every target recognition we provide a unique id that continually increases for identical targets. If you have two surfer targets currently being tracked, for example, one will be 'surfer1', while the other will be 'surfer2'. If you were to track yet another one it would be 'surfer3' and so forth. Our users can choose whether they want to consider these unique ids or not. So you can actually do both scenarios: have the same augmentation on identical targets or have different ones. or any combination thereof even.

Thanks
Eva

 

Login or Signup to post a comment