ARCore | ARCore Course Welcome to the ARCore Course Repo | Augmented Reality library

Finally, I found the solution. According to this answer - https://stackoverflow.com/a/68255789/5709159

it was needed to update this dependency

dependencies { … implementation 'com.google.ar:core:' }

My version was 1.13.0 and the available one was 1.29.0, after the update, the problem is gone

I was able to find a working solution here: https://github.com/chnouman/SceneView

Here are the relevant snippets of code, with some of my adaptations to make it work for .glb files.

I have forked this repo and am working on keyboard input support if anyone's interested in that.

Rendering the object:

At the low level, RealityKit is only responsible for rendering virtual objects and overlaying them on top of the camera frame. If you want to illuminate the real scene, you need to post-process the camera frame.

Here are some tutorials on how to do post-processing: Tutorial1⃣️ Tutorial2⃣️

If all you need is an effect like This , then all you need to do is add a CGImage-based post-processing effect for the virtual object (lights).

More specifically, add a bloom filter to the rendered image(You can also simulate bloom filters with Gaussian blur).

In this way, the code is all around UIImage and CGImage, so it's pretty simple😎

If you want to be more realistic, consider using the depth map provided by LiDAR to calculate which areas can be illuminated for a more detailed brightness.

Or If you're a true explorer, you can use Metal to create a real world Digital Twin point cloud in real time to simulate occlusion of light.

Alpha Blending and Transmission are two different types of transparency – you can combine them, but it's usually unnecessary. Transmission is the newer and more advanced of the two, and it's quite possible that Scene Viewer doesn't support it yet. Alpha Blending is older and should work anywhere.

To use either one separately, you can follow these steps. I'll use Blender's terminology, but hopefully the steps are similar in other software.

Alpha Blend

1. On a Principled BSDF material, set Alpha Mode → Alpha Blend
2. Set Alpha to a value <1, or connect the alpha channel of a base color texture.
3. Export to glTF 2.0

Transmission

1. On a Principled BSDF material, set Alpha Mode → Opaque
2. Set Transmission to a value >0
3. Set Metalness to 0
4. Export to glTF 2.0

You should check the .obj file first, did you export it by yourself? Cause there are many different settings with those kind of files:

1 - maybe your model's faces have 4 vertices (square) instead of 3 (triangles) like the pawn model, 2 - or maybe you didn't set a texture in your .obj file and that create conflict with the shader.

You can just open it with an editor and check how many numbers there are after the "f", if there are 4 number that's the problem, it has square faces, you can open it in Blender and in the export settings check "triangle faces" option. You can also add textures in a couple of minutes, and import them in your project.

I hope this helps you, have a nice day!

AR Core requires some specific hardware to work. You can check the list of supported devices here. No amount of side loading will help because this is a hardware requirement issue. Moreover AR Core is under active development even if you somehow install a version that might work that version will soon be deprecated and you will start getting the popup saying you need to update.

Kindly use a device that is part of supported list or an Emulator that supports this. IMHO it is best to develop using a device that has support from AR Core team.

You can place an anchor on the vertical wall and then attach the ViewRenderable node to that (make it a child of the anchor node object).

It sounds really weird but the Google Pixel 5 doesn't have a ToF sensor for Scene Reconstruction in ARCore. Instead it relies on the wide and ultra-wide lenses for depth estimation (using a Depth-from-Motion algorithm) inside new Depth API. No comments.

I would take a closer look at phones like Samsung Galaxy S20 Ultra. It supports not only Depth API's software Depth-from-Motion algorithm, but also Time-of-Flight hardware depth sensor.

Nonetheless, both gadgets support the same set of features found in ARCore 1.25.

Unfortunately, only one client can use a camera at a time, and both ARCore and CameraX assume they're the only user.

This would require explicit sharing a camera instance between the two, and while I believe ARCore has some provisions for this, I don't believe CameraX is able to use ARCore's interfaces.

So if you need the RGB image, you probably need to figure out how to ask ARCore for it, and not use CameraX at all.