sceneform-android VS Ar-Object-Detection

There's a community-led version of Sceneform that's receiving new features, bug fixes, etc.: https://github.com/SceneView/sceneform-android

Second step, implement an AR sample from the newly maintained github repo for sceneform(previously discontinued from Google and now its being supported by open-source). https://github.com/SceneView/sceneform-android

SceneView/Sceneform: https://github.com/SceneView/sceneform-android

If you are limited by time I'd recommend you to look at Sceneform. It's a deprecated library for easier access to ARCore that will do most of the heavy lifting. Read the official documentation from Google first, then check out the community-maintained version on GitHub, which is (for he most part) the same. The readme has a section on the differences.

I want to use sceneform maintained .. In the demo app (AR Environment Lights) there's an option that changes "skybox" from camera to color. I need to implement a similar thing for a university project and I can't quite find out how they did it. I want the AR to happen without the camera preview..

are you using pure ARCore or Sceneform? Have a look at this

Sceneform was depreciated, there's a maintained version if you're looking for something to help you develop android side: Sceneform Android

For native Android developers. We are a small community hardly working everyday on this: https://github.com/ThomasGorisse/SceneformMaintained Sceneform Maintained is an ARCore Android SDK with Google Filament as 3D engine. This is the continuation of the archived Sceneform

For Sceneform Maintained, just download this sample project: https://github.com/ThomasGorisse/SceneformMaintained/tree/master/samples/gltf

We need sponsoring: https://github.com/ThomasGorisse/sceneform-android-sdk

Based on what I understood from the last part, that the camera deals with bitmaps, and ArCore I don't know how it deal with camera, my knowledge starts from the hit result. So search for two days here and there, I found Kashif-E/Ar-Object-Detection, but as mush I try I cant run it on any model.