@Jeremy Face service or Face API is a separate service or API that could be created from Azure portal before MS decided to restrict access to the face API and other sensitive AI services to support the responsible AI principles to avoid any misuse. The addition of face api to the computer vision documentation is inline with the changes in vision pillar of cognitive services where the vision studio will offer all vision related APIs under one portal. The current documentation still recommends using a separate face resource since it requires special access which is submitted through a form.
Computer vision Celebrity Recognition API also requires special access since it comes under limited access feature of computer vision.
To summarize, due to the limited access features in both the services face API is still a separate resource since almost all of the features of the API as sensitive in nature whereas computer vision has few features which need special access. This ensures that users who want to use OCR, Image analysis and spatial analysis features get access to the API without any special access.
If an answer is helpful, please click on or upvote
which might help other community members reading this thread.