How can we help you today?

Start a new topic
Discussions Wikitude SDK (Android, iOS, UWP) 3D Model Questions or Problems

Use Wikitude as model rendering without GeoPosition

Use Wikitude as model rendering without GeoPosition
11 CommentsSorted by Oldest First

Hi all, in which way I can use the SDK for render model into the tablet display without the GPS position?

My goal is to display my model in the middle of display with the ability to move/scale it with the gesture from JS.

Anyone have some example for me?

 

Thanks

Hello,

the Wikitude SDK doesn't support the use of a 3d Model in such a way.

Regards,
Christian

I understand it's not the proper use for this SDK.

However I made some test and I manage to render 3d model and manipulate the object by using JS gesture.

Our issue are related to the camera/device position into the world, there's a way to know or manually change this position?

Obviously at the moment we work with sensors off.

We think that if we solve the position of a camera we are also able to manage the position of the model into the device display.

 

I'm looking for stuff because the second step of this project is the AR parts so Wikitude is the natural choice for this job.

 

Thanks

For such an operation you'd need the device orientation matrix which is not available in JS. And even if you had the matrix it would be quite a hack to use it for such a calculation. Furthermore it wouldn't even perform well since you'd need to fake/reset the position within a very short time period. A support of this feature by our SDK would be the best solution for your problem of course. I'll forward your feature request. Sorry for this limitation.

Regards,
Christian

Is there any progress or plans on that feature? It would be a really important feature for us because in Titanium that we are using there is not a good cross platform solution to render a 3D model, so your work on the renderer would really help us

Regards,

Nikos

Not such a feature as described above. Have a look at the 'snap to screen'-feature in our SDK examples. Sounds like this is what you are looking for.

I checked the snap to screen feature but this still requires the model to be binded to a trackable right? What we would like to have is the ability to render a 3D model directly without using the location or camera of the device...

The core feature of our product is augmented reality. The Wikitude SDK is not a 3d rendering engine. There are products out there which are specialized on that like three.js for example.

Hi Enrico

What you did seems fantastic!

I was interested in something similar, can you share with us how did you manage to do something like that?

A code snippet would be highly appreciated.

Regards,

Vishy

Wrong channel, Vishy?

Hii, not sure what you mean, but am not using Wikitude for this now, am directly doing what I wanted via Unity.
Login or Signup to post a comment
More topics in 3D Model Questions or Problems
See all 403 topics
© 2023 Wikitude, a Qualcomm company