XRViewerPose

In addition to the properties inherited from XRPose, XRViewerPose includes the following:

views Read only

An array of XRView objects, one for each viewpoint on the scene which is needed to represent the scene to the user. A typical headset provides a viewer pose with two views whose eye property is either left or right, indicating which eye that view represents. Taken together, these views can reproduce the 3D effect when displayed on the XR device.

The XRViewerPose object is used to describe the state of a viewer of a WebXR scene as it's tracked by the user's XR hardware. The viewer may be the virtual representation of the user, or it may represent another device or interface which may serve as the source of a position and orientation that make up a view upon the scene. For example, every player in a MMORPG might have an instance of XRViewerPose to provide a way to calculate what they can see; if the game provides a mechanism that tells the player if another player sees them, or that they see another player, this information becomes crucial.

An XRViewerPose is always obtained and referenced relative to an existing XRReferenceSpace. This ensures that positions and orientations are reported using the expected relative coordinate system.

To render a scene using the XRViewerPose representing the user's head, one would iterate over the views in the views array, rendering them one after another. By calling viewport() on the WebGL context, specifying the XRView as input, you can get the viewport to use when rendering in order to draw the frame for that eye into the correct part of the drawing surface.

Also, when rendering the scene for spectators or other players in a multiplayer game, the transform of the XRViewerPose can be used to determine both placement and facing direction of the other players in the game, so that they can be drawn in the correct place with the correct facing.

The viewer's pose for the animation frame represented by XRFrame can be obtained by calling the frame's getViewerPose() method, specifying the reference space in which the origin's position should be computed. The returned XRViewerPose tells you where the viewer is and what direction they're facing at the time at which the frame takes place.

In this example—part of the code to render an XRFrame, getViewerPose() is called to get an XRViewerPose using the same reference space the code is using as its base reference space. If a valid pose is returned, the frame is rendered by clearing the backbuffer and then rendering each of the views in the pose; these are most likely the views for the left and right eyes.

const pose = frame.getViewerPose(xrReferenceSpace);

if (pose) {
  const glLayer = xrSession.renderState.baseLayer;

  gl.bindFrameBuffer(gl.FRAMEBUFFER, glLayer.framebuffer);
  gl.clearColor(0, 0, 0, 1);
  gl.clearDepth(1);
  gl.clear(gl.COLOR_BUFFER_BIT, gl.DEPTH_BUFFER_BIT);

  for (const view of pose.views) {
    const viewport = glLayer.getViewport(view);
    gl.viewport(viewport.x, viewport.y, viewport.width, viewport.height);

    /* render the scene for the eye view.eye */
  }
}

Passing each view to getViewport() returns the WebGL viewport to apply in order to cause the rendered output to be positioned correctly in the framebuffer for rendering to the corresponding eye on the output device.

This code is derived from Drawing a frame in our "Movement and motion" WebXR example. You can see more context and see much more on that page.

• WebXR Device API
• Movement, orientation, and motion
• XRPose and XRView