Start a new topic

placing of 3D model without tracker or geolocation

placing of 3D model without tracker or geolocation


Hi Team Wikitude,

I have created a AR.Model

Now I just want to give some location to this model without a tracker or geo location.

The location can be hardcoded and then allow user to rotate/pan/zoom the model.

I could not find any such sample code in the SDK samples.

Can you please help me out here.

Highly appreciate your inputs.

Regards,

Vishy

Hi Vishy,

you need to have at least a geo-location (relative location could work as well) or a tracker to attach a 3D model to it. The rest of your requirements is captured in the 3D sample and documentation. 

 

Hi Philipp,

Thanks for the reply.

Suppose I use to RelativeLocation achieve this requirement

If I pass location as null, it says it will use current user location, does this mean that if I keep it null I wont need GPS and network on?

So if i specify

AR.RelativeLocation(null, 12, 0, -5);

 

Will the model be in 12 m in front of the phone, at the center (since easting is 0) and 5 meters below the phone location?

Another query I had was(pardon for mixing up the queries, If required I will create a new thread for this)

Can I give user option to zoom/pan/rotate the model using hand touch gestures like pinch and swipe?

I could not find such an sample script for this in the examples (I may be wrong)

If this is possible could you help with a code snippet or point me to the relevant example.

Thanks & Regards,

        Vishy

Hi,

you will need the sensors on for that to work. The 3D model will be placed 12 meters aways in east direction of your current location. 

Did you have a look at the 3D sample I referenced before - it includes a section called Snap to screen, which has pan/rotate and pinch-to-zoom functionality included. 

12 m east or north?
 

Hi Phillip,

The Snap to screen example is a good example to learn about pan/zoom/rotate.

(I wanted to mark your reply as answer, but somehow cant see the option to do that!)

Somehow, in the example the car only rotates, there is no pan or zoom. Not sure if that option has been given.

The other problem I am facing is that the logic for these examples are in java script which is very difficult for me to debug and try new things.

Is it not possible for this code to be available in java directly?

This would help us in exploring the different member functions very fast and debug.

Thanks & Regards,

   Vishy

 

 

Hi,

you are right, that the example on Android is lacking pinch-to-zoom. It seems the Android web view interprets that differently.  Pan isn't implemented in this case as wanted to keep the model in the center.

JavaScript is currently (SDK 4.1 or earlier) the only way to interact with the augmented reality scene. The SDK 5 will have a Native API, where you can code directly in Java. However, this will not provide any means for rendering initially.

Is the Android web view issue something we can handle by modifying the code or is there a change required in the SDK?

Would it be possible to get an example js which pan/zoom/rotates the model based on hand gestures?

Need some hand holding for this and this would be a very good example for lot of developers.

Would highly appreciate any such sample script.

Hi,

the Android issue is something you can fix on your side as well. It is just how we capture the touch events in this sample. At the moment I can't help you with a sample, maybe someone from the community can.

 

Thanks for the update Philipp.

If anyone in the community can help me out in this, do let me know.

I am open to do some try outs myself too if we have a pseudo code available.
Login or Signup to post a comment