Hello,
Does anyone have a recipe on how to raycast VNFaceLandmarkRegion2D points obtained from a frame's capturedImage?
More specifically, how to construct the "from" parameter of the frame's raycastQuery from a VNFaceLandmarkRegion2D point?
Do the points need to be flipped vertically? Is there any other transformation that needs to be performed on the points prior to passing them to raycastQuery?
Hello @jimbang,
I see, is there a specific reason that you want to do this with the rear camera rather than the front? The front camera will enable you to take advantage of ARFaceAnchor.
I don't have a strong recommendation for doing this with the rear camera. You can use Vision to detect 2D face landmarks, and then use depth data to get world points for those landmarks, but the measurements from that approach are likely to not meet your requirements.
--Greg