Custom 3D character model for RealityKit Motion Capture

At WWDC 2019 and in the demo project body tracking data is applied to the robot model robot.usdz. I want to use a custom 3D character model. But there is no specification how to rig the character model in order to be compatible with the RealityKit implementation. The only information I found is the statement at WWDC "the model must follow the same structure of robot.usdz". And I can view the joints when I open the robot model with XCode.


Since Apple does not provide a specification about the 3D character model. Does anyone know

  • - what is required (i.e. what does RelalityKit read) and what must be the same (e.g. naming, orientation)?
  • - a rig tool which by default provides the required set of joints?
  • - a workflow how to set up a character model? (e.g. import robot.usdz with Mulitverse in Maya, replicate on a new model, export as fbx, convert with RealityConverter to usdz)

Accepted Reply

Hello,


Apple has recently provided the specifications in the following links :

https://developer.apple.com/documentation/arkit/rigging_a_model_for_motion_capture

https://developer.apple.com/documentation/arkit/validating_a_model_for_motion_capture


Good luck !

Replies

Hello,


Apple has recently provided the specifications in the following links :

https://developer.apple.com/documentation/arkit/rigging_a_model_for_motion_capture

https://developer.apple.com/documentation/arkit/validating_a_model_for_motion_capture


Good luck !