I made the switch to the tapo H500 hub. It is a great product but it's far from ready to take over my DVR. There are many limitations that still makes me hold the full move, like the limitation to continuously record only 4 cameras. Face recognition is good but is very limited. And because of this I have a question:

After I go to the unrecognized faces and categorize them properly to persons, does the hub retrain the face recognition model? Looks like no matter what I do to improve recognition, it doesn't get any better at recognizing people's faces.

In comparison, Netatmo cameras do this by default: the more you tag people's names from the events, the better the recognition is.

Would be good to have an official answer to this capability from TpLink.