Resolution and positional accuracy are very poor. It’s more like ‘an approximate bag of water detector’.

Gait analysis is complete fiction. Especially with a non-visual approach like this.

Given the number of gait analysis publications over several decades using varying techniques, can you recommend a good review article disproving all of them?

Given the number of publications about curing <pick your uncured disease> over several decades using varying techniques, can you recommend a good review article disproving all of them?

Answer: no need, if it had been cured, it would be cured. And it is not.

My point being that many publications saying "towards X" may mean that we are making some progress towards X, but they don't mean at all that X is possible.

I don’t think anyone has ever tried to publish something disproving all of the gait analysis claims. That would be an odd sort of thing. But I have not seen anything come to something that we could call productized and reliable. It’s relatively easy to publish theoretical papers. Much harder to show it working reliably in the wild.

If you can do that you can infer when someone is home or away.