SceneView Android

3D and AR Android Composable and View with Google Filament and ARCore

This is a Sceneform replacement in Kotlin

Features

• Use sceneview dependency for 3D only or arsceneview for 3D and ARCore.
• Compose: Use the Scene or ARScene @Composable
• Layout: Add the <SceneView> or <ArSceneView> tag to your layout or call the ArSceneview(context: Context) constructor in your code.
• Requesting the camera permission and installing/updating the Google Play Services for AR is handled automatically in the ArSceneView.
• Support for the latest ARCore features (the upcoming features will be integrated quicker thanks to Kotlin).
• Lifecycle-aware components = Better memory management and performance.
• Resources are loaded using coroutines launched in the LifecycleCoroutineScope of the SceneView/ArSceneView. This means that loading is started when the view is created and cancelled when it is destroyed.
• Multiple instances are now possible.
• Much easier to use. For example, the local and world position, rotation and scale of the Node are now directly accessible without creating Vector3 objects (position.x = 1f, rotation = Rotation(90f, 180f, 0f), scale = Scale(0.5f), etc.).

Architecture

Youtube

Dependency

app/build.gradle

• 3D (Filament included)

dependencies {
    // 3D only
    implementation 'io.github.sceneview:sceneview:1.0.4'
}

API Reference

• AR (Filament + ARCore included)

dependencies {
    // 3D and ARCore
    implementation 'io.github.sceneview:arsceneview:0.9.5'
}

API Reference

Usage

3D

• Compose

@Composable
fun ModelScreen() {
    val nodes = remember { mutableStateListOf<Node>() }

    Box(modifier = Modifier.fillMaxSize()) {
        Scene(
            modifier = Modifier.fillMaxSize(),
            nodes = nodes,
            onCreate = { sceneView ->
                // Apply your configuration
            }
        )
    }
}

• Layout

<io.github.sceneview.SceneView
    android:id="@+id/sceneView"
    android:layout_width="match_parent"
    android:layout_height="match_parent" />

AR

• Compose

@Composable
fun ARScreen() {
    val nodes = remember { mutableStateListOf<ArNode>() }

    Box(modifier = Modifier.fillMaxSize()) {
        ARScene(
            modifier = Modifier.fillMaxSize(),
            nodes = nodes,
            planeRenderer = true,
            onCreate = { arSceneView ->
              // Apply your configuration
            },
            onSessionCreate = { session ->
              // Configure the ARCore session
            },
            onFrame = { arFrame ->
              // Retrieve ARCore frame update
            },
            onTap = { hitResult ->
              // User tapped in the AR view
            }
        )
    }
}

• Layout

<io.github.sceneview.ar.ArSceneView
    android:id="@+id/sceneView"
    android:layout_width="match_parent"
    android:layout_height="match_parent" />

3D Model Viewer

Youtube

ModelNode(
    position = Position(x = 0.0f, y = 0.0f, z = -4.0f),
    rotation = Rotation(y = 90.0f),
    scale = Scale(0.5f)
)

Parameters

• position The node position to locate it within the coordinate system of its parent
  Default is Position(x = 0.0f, y = 0.0f, z = 0.0f), indicating that the node is placed at the origin of the parent node's coordinate system.
  
• rotation The node orientation in Euler Angles Degrees per axis from 0.0f to 360.0f The three-component rotation vector specifies the direction of the rotation axis in degrees. Rotation is applied relative to the node's origin property.
  Default is Rotation(x = 0.0f, y = 0.0f, z = 0.0f), specifying no rotation.
• scale The node scale on each axis
  Reduce (scale < 1.0f) / Increase (scale > 1.0f)

AR Model Viewer

Youtube

ArModelNode(
    placementMode = PlacementMode.BEST_AVAILABLE, 
    hitPosition = Position(0.0f, 0.0f, -2.0f),
    followHitPosition = true,
    instantAnchor = false
)

Parameters

• placementMode Define the AR Placement Mode depending on your need
  You can change it to adjust between a quick (PlacementMode.INSTANT), more accurate (PlacementMode.DEPTH), only on planes/walls (PlacementMode.PLANE_HORIZONTAL, PlacementMode.PLANE_VERTICAL, PlacementMode.PLANE_HORIZONTAL_AND_VERTICAL) or with auto refining accuracy placement (PlacementMode.BEST_AVAILABLE).
  The hitTest, pose and anchor will be influenced by this choice.
• hitPosition The node camera/screen/view position where the hit will be made to find an AR position
  Until it is anchored, the Node will try to find the real world position/orientation of the screen coordinate and constantly place/orientate himself accordingly followHitPosition is true.
  The Z value is only used when no surface is actually detected or when followHitPosition and instantAnchor is set to false or when instant placement is enabled.
• followHitPosition Make the node follow the camera/screen matching real world positions
  Controls if an unanchored node should be moved together with the camera.
  The node position is updated with the realtime ARCore pose at the corresponding hitPosition until it is anchored (isAnchored) or until this this value is set to false.
  • While there is no AR tracking information available, the node is following the camera moves so it stays at this camera/screen relative position but without adjusting its position and orientation to the real world
  • Then ARCore will try to find the real world position of the node at the hitPosition by looking at its hitTest on each onArFrame.
  • In case of instant placement disabled, the z position (distance from the camera) will be estimated by the AR surface distance at the (x,y).
  • The node rotation will be also adjusted in case of PlacementMode.DEPTH or depending on the detected planes orientations in case of PlacementMode.PLANE_HORIZONTAL, PlacementMode.PLANE_VERTICAL, PlacementMode.PLANE_HORIZONTAL_AND_VERTICAL
• instantAnchor Anchor the node as soon as an AR position/rotation is found/available
  If true, the node will be anchored in the real world at the first suitable place available

AR Placement Mode

Youtube

Choose how an object is placed within the real world

• DISABLED Disable every AR placement preview and handle it by yourself (onTap, onAugmentedFace, onAugmentedImage
• PLANE_HORIZONTAL Place and orientate nodes only on horizontal planes
• PLANE_VERTICAL Place and orientate nodes only on vertical planes
• PLANE_HORIZONTAL_AND_VERTICAL Place and orientate nodes on both horizontal and vertical planes
• DEPTH Place and orientate nodes on every detected depth surfaces. Not all devices support this mode. In case on non depth enabled device the placement mode will automatically fallback to PLANE_HORIZONTAL_AND_VERTICAL.
• INSTANT Instantly place only nodes at a fixed orientation and an approximate distance. No AR orientation will be provided = fixed +Y pointing upward, against gravity. This mode is currently intended to be used with hit tests against horizontal surfaces.
• BEST_AVAILABLE Place nodes on every detected surfaces. The node will be placed instantly and then adjusted to fit the best accurate, precise, available placement.

Parameters

• instantPlacementDistance Distance in meters at which to create an InstantPlacementPoint. This is only used while the tracking method for the returned point is InstantPlacementPoint.
  Default: 2.0f (2 meters)
• instantPlacementFallback Fallback to instantly place nodes at a fixed orientation and an approximate distance when the base placement type is not available yet or at all.

Load a glb/glTF Model

Asynchronously

modelNode.loadModelAsync(
    context = context,
    lifecycle = lifecycle,
    glbFileLocation = "models/mymodel.glb",
    autoAnimate = true,
    autoScale = false,
    centerOrigin = null,
    onError = { exception -> },
    onLoaded = { modelInstance -> }
)

Within a Coroutine Scope

lifecycleScope.launchWhenCreated {
    val modelInstance = modelNode.loadModel(
        context = context,
        glbFileLocation = "https://sceneview.github.io/assets/models/MaterialSuite.glb",
        autoAnimate = true,
        autoScale = true,
        centerOrigin = Position(x = 0.0f, y = 0.0f, z = 0.0f),
        onError = { exception -> }
    )
}

Parameters

• lifecycle Provide your lifecycle in order to load your model instantly and to destroy it (and its resources) when the lifecycle goes to destroy state
  Passing null means the model loading will be done when the Node is added to the SceneView and the destroy will be done when the SceneView is detached.
• modelFileLocation The model glb/gltf file location
  • A relative asset file location (models/mymodel.glb)
  • An Android resource from the res folder (context.getResourceUri(R.raw.mymodel)
  • A File path (Uri.fromFile(myModelFile).path)
  • An http or https url (https://mydomain.com/mymodel.glb)
• autoAnimate Plays the animations automatically if the model has one
• autoScale Scale the model to fit a unit cube so it will better fit your SceneView
• centerOrigin Center point origin position within the model
  Float cube position values between -1.0 and 1.0 corresponding to percents from model sizes.
  • null = Keep the origin point where it was at the model export time
  • Position(x = 0.0f, y = 0.0f, z = 0.0f) = Center the model horizontally and vertically
  • Position(x = 0.0f, y = -1.0f, z = 0.0f) = center horizontal | bottom
  • Position(x = -1.0f, y = 1.0f, z = 0.0f) = left | top
  • ...
• onError An exception has been thrown during model loading

AR Cloud Anchors

Youtube

Sources

sceneView.cloudAnchorEnabled = true

// Host/Record a Cloud Anchor
node.onAnchorChanged = { node: ArNode, anchor: Anchor? ->
    if(anchor != null) {
        node.hostCloudAnchor { anchor: Anchor, success: Boolean ->
            if (success) {
                // Save the hosted Cloud Anchor Id
                val cloudAnchorId = anchor.cloudAnchorId
            }
        }
    }
}

// Resolve/Restore the Cloud Anchor
node.resolveCloudAnchor(cloudAnchorId) { anchor: Anchor, success: Boolean ->
    if (success) {
        node.isVisible = true
    }
}

AR Depth/Objects Occlusion

Youtube

sceneView.isDepthOcclusionEnabled = true

This will process the incoming ARCore DepthImage to occlude virtual objects behind real world objects.
If the AR Session is not configured properly the standard camera material is used.
Valid Session.Config for the Depth occlusion are Config.DepthMode.AUTOMATIC and Config.DepthMode.RAW_DEPTH_ONLY
Disable this value to apply the standard camera material to the CameraStream.

AR Geospatial API

Youtube

Follow the official developer guide to enable Geospatial in your application. For configuring the ARCore session, you just need to enable Geospatial via ArSceneView.

• Enable Geospatial via ArSceneView

arSceneView.geospatialEnabled = true

• Create an Anchor

val earth = arSceneView.session?.earth ?: return
if (earth.trackingState == TrackingState.TRACKING) {
    // Place the earth anchor at the same altitude as that of the camera to make it easier to view.
    val altitude = earth.cameraGeospatialPose.altitudeMeters - 1
    val rotation = Rotation(0f, 0f, 0f)
    // Put the anchor somewhere around the user.
    val latitude = earth.cameraGeospatialPose.latitude + 0.0004
    val longitude = earth.cameraGeospatialPose.longitude + 0.0004
    earthAnchor = earth.createAnchor(latitude, longitude, altitude, rotation)
}
// Attach the anchor to the arModelNode.
arModelNode.anchor = earthAnchor

Camera Permission and ARCore install/update/unavailable

ArSceneView automatically handles the camera permission prompt and the ARCore requirements checks. Everything is proceed when the attached view Activity/Fragment is resumed but you can also add your ArSceneView at any time, the prompt will then occure when first addView(arSceneView) is called.

If you need it, you can add a listener on both ARCore success or failed session creation (including camera permission denied since a session cannot be created without it)

• Camera permission has been granted and latest ARCore Services version are already installed or have been installed during the auto check

sceneView.onArSessionCreated = { arSession: ArSession ->
}

• Handle a fallback in case of camera permission denied or AR unavailable and possibly move to 3D only usage

sceneView.onArSessionFailed = { exception: Exception ->
    // If AR is not available, we add the model directly to the scene for a 3D only usage
    sceneView.addChild(modelNode)
}

The exception contains the failure reason. e.g. SecurityException in case of camera permission denied

Customizing the instructions

• The default instruction nodes have a ViewRenderable with a TextView or ImageView
• The text and images of the instruction nodes can be overridden at the resource level (in the strings.xml file and drawable directory of your project).
• Custom instruction nodes can have an arbitrary number of child nodes with ModelRenderables and ViewRenderables. It is even possible to play animation for a ModelRenderable if it is defined in a .glb file or a video using the VideoNode
• The infoNode can have one of the following values depending on the ARCore features used and the current ARCore state: searchPlaneInfoNode, tapArPlaneInfoNode and augmentedImageInfoNode. Alternatively, it is possible to create your own instruction nodes.
• The SearchPlaneInfoNode displays messages related to the ARCore state. For example, Searching for surfaces..., Too dark. Try moving to a well-lit area, Moving too fast. Slow down, etc.
• The TapArPlaneInfoNode displays a message that helps users to understand how an object can be placed in AR when no objects are currently present in the scene.
• The AugmentedImageInfoNode displays a frame with white corners when no augmented image is currently tracked.

💡 Idea for future: when access to the flashlight is finally available with the ARCore shared CameraManager, it will be great to add a button to the SearchPlaneInfoNode to enable the flashlight when there isn't enough light.

Why have we included the Kotlin-Math library in SceneView?

Earlier versions of OpenGL had a fixed rendering pipeline and provided an API for setting positions of vertices, transformation and projection matrices, etc. However, with the new rendering pipeline it is required to prepare this data before passing it to GLSL shaders and OpenGL doesn't provide any mathematical functions to do that.

It is possible to implement the required functions yourself like in Sceneform or use an existing library. For example, C++ supports operator overloading and benefits from the excellent GLM library that allows to use the same syntax and features as GLSL.

We use the Kotlin-Math library to rely on a well-tested functions and get an advantage of using Kotlin operators for vector, matrix and quaternion operations too.

Migration from Sceneform

You will have a little work to do if you are using the ArFragment in Sceneform. However, there is the Deprecated.kt file to help you with the migration.

Using the migration suggestions

1. Remove the Sceneform import for the class you want to migrate.
2. Import this class from the io.github.sceneview.ar package.
3. Use Alt+Enter/the light bulb icon to view and apply the suggestions for replacing the deprecated method calls.

After the migration you should get cleaner code and all of the benefits described in the Features section 🎉

Requesting the camera permission and installing/updating the Google Play Services for AR

This is handled automatically in the ArSceneView. You can use the ArSceneView.onArSessionFailed property to register a callback to be invoked when the ARCore Session cannot be initialized because ARCore is not available on the device or the camera permission has been denied.

Instructions for AR

The InstructionsController in the BaseArFragment has been replaced with the Instructions in the ArSceneView.

The Instructions use a Node that is a part of the scene instead of a View, as opposed to the InstructionsController. This provides more flexibility for customizing the instructions. The Instructions have the main Node that can be accessed through the Instructions.infoNode property.