Layering ML on top of AR apps extends their usefulness.
AR technology often uses SLAM (simultaneous localization and mapping):
In conclusion.
Because of the technological progress made in the last couple of years, AR apps are exceeding expectations in terms of what they can do.
Since the camera image is always available, these apps also provide a unique opportunity for using visual data to enrich the experience: Enabling the app to read text, find objects or detect anomalies, for example. These features can be powered by a custom machine learning layer built on top of the core tech.
See the full story here: https://towardsdatascience.com/enhancing-ar-with-machine-learning-9214d2da75a6