Hi Nate1244. Thank you for the question. Short: No. Long: You can access a lot of the ModelEntities of the scene. Then you can add either anchors to the scene anchors with your custom objects which can be queries if the recognized window is actually a window. It would require user input but acc would be 100%. For other more advanced solutions you would need the Vision framework, OpenCV or similar. There are a lot of demo projects on Github. One is a more advanced app which actually shows textual overlays. If you need more info feel free to ask. Best regards, Mladen

Hi sducouedic. Thank you for the question. There is only one way which I have found that actually works, and will be working. You will have to map the CaptureRoom to your own structure which conforms to Codable. You can make a class or struct out of it. The important thing is to document the RoomPlan version in your mapped class-stuct. This way you can start working with versions in the future. If you need any more info feel free to ask. Best regards, Mladen

Hi. Thank you for this question. You don't need to convert a usdz to JSON in your example. You can convert the CapturedRoom object to JSON with this free library: https://github.com/fhstp/swift-roomplan-jsonexport.git I have tried to find a way to convert usdz to JSON but have not found any libs that can do this. The above works for CapturedRoom to JSON. Best regards, Mladen

Me2. The same source code works on an iPhone with LiDAR.

`Copy usdpython from /Applications to ~/Developer as follows - $ cp -R /Applications/usdpython ~/Developer/ Then add the following to your .bash_profile` This did not work.