![]() ![]() ![]() |
|
jonathan18: So both mine are the pan/scan model. Given they’re both outside, I thought this would make them more flexible in terms of coverage. In practice, the front one is run stationary, as the angle of view is already wide enough to see both our front driveways. (Note, I think this model already has a slightly wider angle lens). The rear I run panning on a loop to cover both the back door and the sleep out door. The problems with that are the restrictions on the length one has to play with auto panning (four positions over 30 secs, I think), the noise at night (it’s bracket is screwed to the outside wall, and that amplifies it), and the current inability to turn pan/scan on/off with shortcuts (so can’t turn it off automatically at night).
Doing it again I’d have gone with three standard cameras, using two to cover the back of the house.
Thanks, that was helpful. I've just ordered 2 x standard cameras and 2 x 32GB SD cards from Amazon $114. Looking forward to retiring my aging D-link.
dafman: I will be using one Wyze to view outside through a window. Anyone else doing this? I read somewhere that I will need to tape over the night vision lights as these will reflect on the window otherwise?
I recall reading that, but I wonder if it's not necessary now (perhaps a firmware update?) given there's an option to disable IR lights in the advanced settings - this is what the app reports:
dafman: I will be using one Wyze to view outside through a window. Anyone else doing this? I read somewhere that I will need to tape over the night vision lights as these will reflect on the window otherwise?
I've ordered 2 from Amazon and it just arrived for the weekend. I'm using them to view outside through a window too. There is an option to turn off the IR now so that makes it easier. Works quite well but of course in the dark, you're not going to see much unless an external sensor light turns on. But it will avoid the reflection off the glass. Still doing some more testing and fine-tuning. But happy so with how it works so far.
https://youtu.be/IHaPxGRNcug
To put it nicely, this person detection is a work in progress.
As it happens I was looking at this this morning - of the five motion detections recorded today, one was a car light and four were catching me going on/returning from my morning walk. Not one of those four videos with me clearly in them were identified by this AI detection. It does better in daylight, but is still only picking up a proportion of those it should. Doesn't seem to be a problem with false positives, but I suggest it's not at a point that it's worth relying on it to distinguish between people and other.
The app (perhaps only the beta version?) has the ability to share detection videos with Wyze, allowing you to identify whether a person is in the video, so clearly their AI is still at school and yet to graduate.
|
![]() ![]() ![]() |