Short, periodic zoom in&out to face on AI person identicifation

If the camera recognizes an object as a person it would be helpful to get a short (1-2 second), automated, zoom-in on the head every 10-20 seconds (user setting) or so. This image could just be inserted into the video stream. Even if AI can’t identify the person from stored identity data, the user can pause the playback and hopefully get a better read on the person’s identity.

Check out the “Smart Focus” on the Wyze Cam V3 Pro camera. Very similar to what you are mentioning.