Based on what you mentioned all of those are also available in ARKit or Vision API: plane detection (vertical and horizontal), anchors (both local and geolocate

Based on what you mentioned all of those are also available in ARKit or Vision API: plane detection (vertical and horizontal), anchors (both local and geolocated), 3d scene reconstruction, 3d object reconstruction, custom planar markers, qrcode & barcodes, 3d human pose skeleton, 3d hand skeleton, face landmarks mesh, world tracking (SLAM), text detection. Haven't checked Meta APIs but it doesn't look for me as far ahead of Apple.