In ARKit 3, person segmentation with depth and body detection frame semantics seem to be mutually exclusive. It crashes returning the error: "This set of frame semantics is not supported on this configuration." Is this still the case in ARKit 4?
Is it possible to use people occlusion and motion capture at the same time in ARKit 4
Hello,
You should always check if a particular combination of frame semantics is supported on your current device, and for a particular configuration, using supportsFrameSemantics(_:).
The following returns false:
You should always check if a particular combination of frame semantics is supported on your current device, and for a particular configuration, using supportsFrameSemantics(_:).
The following returns false:
Code Block ARWorldTrackingConfiguration.supportsFrameSemantics([.bodyDetection, .personSegmentationWithDepth])
Yes, this combination is still unsupported.
Using an Arkit device as a camera, I can body track collision myself to punch a 3d model next to me. But my arms behind the model will show up infront of the model. So in order to properly "fight" a 3d model, I need occlusion AND body tracking.
Would love to get word on when we can use both at the same time!!!!!!!!
Would love to get word on when we can use both at the same time!!!!!!!!
I am also looking forward for these two features simultaneously. This will really open new doors. Is there some workaround? As far as I tried to figure out - no. However, I would be pleased to know that I'm wrong.
Yes, this can be done using a Reality Scene.