Wikitude

Hmm, hard to tell without seeing more of the project. Did you use render textures in your approach?

You could definitely try using a fullscreen camera and just overlaying it with an overlay with your desired "cutout". Not the most performant way I admit, but the fastet achievable way for sure.

Kind regards,
Gökhan

Hi Leo,

Thats unfortunately not possible since the background camera is generated on runtime and is a hidden object in the hierarchy in our professional edition SDK. In expert edition we introduced the CameraFrameRenderer gameObject therefore.

Nevertheless, there is a workaround to make it visible in professional edition. Just add this script to the WikitudeCamera object.

using System.Collections;
using System.Collections.Generic;
using UnityEngine;

public class BackgroundCameraEnable : MonoBehaviour
{
    public Camera BackgroundCamera;

    void Update() {
        if(!BackgroundCamera) {
            BackgroundCamera = gameObject.transform.Find("BackgroundCamera").GetComponent<Camera>();
            BackgroundCamera.gameObject.hideFlags = HideFlags.None;
        }
    }
}

It should instantly find the object and "unhide" it on entering play mode. You could then use the camera component to resize and modify the camera to your likings.

Kind regards,

Gökhan

Hi Leo,
Could you share, how you are using the Input Plugins API?

Kind regards,

Gökhan

Hi Leo,

i did a quick test and summarized what you can do in order to get a scaled, aspect ratio changed background + ar camera view:

Create two render textures. One will hold the AR Cameras output ("WikitudeCamera") and a second one for the background camera.

They are set up like this in my case. Sizes of the render textures should be set during runtime.

You have to set the WikitudeCamera camera components clear flags to "Solid Color" with a transparent color by script on runtime, because it will be set to "Depth Only" on startup. But for the render textures you need this to be set. You can modify the BackgroundCameraEnable script i posted before (which has to be added to the WikitudeCamera gameobject). This way you can get access to both camera components on runtime.

Also add this two renderTextures as public variables to the script. Create your UI and add two RawImage UI elements to the canvas. I did it like this for instance (keep in mind -> the background camera image has to be behind the AR camera or in this case higher in the hierarchy):

The Raw Images will hold the render textures of the cameras. I set their sizes to their native sizes (by script) and scaled them down to 0.5. The Box element wrapping the raw images has a Rect Mask 2D component, which will cut off everything outside of its borders. The size of the box was set to 500 by 500 in this case, as an example.

Then you should be able to get from this:

toooooooooo this:

You can change the shape of the box anyway you want and position it anywhere you want. Keep in mind to alter the  render texture and ui sizes on device orientation changes.

Let me know if this solution works for you!

Kind regards,
Gökhan

Hi Leo,

Thats probably because the Render Texture for the background camera wasn't set. I would check for this in the update loop and apply the changes if needed. Here a code snippet (untested):

...

void Update() {
   if (backgroundCamera.renderTexture == null) {
      backgroundCamera.renderTexture = backgroundCameraRenderTexture;
   }

   if (backgroundCamera.clearFlags != CameraClearFlags.SolidColor) {
      backgroundCamera.clearFlags = CameraClearFlags.SolidColor;
      backgroundCamera.backgroundColor = new Color(0,0,0,0);
   }
   ...
}

...

Kind regards,
Gökhan

Hi Leo,

I guess you also have to adjust the inputs position and send the adjusted position to the Tracker where you would have touched, if the view would have been fullscreen. Hmmm, let me come of with some pseudo code (not tested) for you...

if (_isTracking && Input.GetMouseButtonUp(0)) {
   if (ARCameraViewRect.Contains(Input.mousePosition) {
      Vector2 adjustedPosition = new Vector2();
      adjustedPostition.x = Input.mousePosition.x - ARCameraViewRect.position.x * (Screen.width / ARCameraViewRect.size.width);
      adjustedPostition.y = Input.mousePosition.y - ARCameraViewRect.position.y * (Screen.height / ARCameraViewRect.size.height);  
    
     Tracker.ConvertScreenCoordinate(adjustedPostition);
   }
}

Kind regards,
Gökhan