The technology wouldn’t work. There’s too much variance in things like the angle the camera is viewing from versus glare from the windshield to get a reliable corpus of data to train from. Even then, the seatbelt isn’t super visible, it’s easily obscured by shadow or body shape or even the colour of your clothing.
The false positive rate would probably match the true positive rate - which is to say it’s a total coin toss and not something you want to rely on for enforcement
In fairness though, I think that’s probably the same for inattentive driving. You probably could do some video processing instead of simple image classification, but that’s a massive increase in complexity for dubious gain.
Yes: speeding and inattentive driving, especially use of phone.
No: seatbelts and broken taillights.
Why not seatbelts? They’re mandatory as well. In for an inch, in for a mile.
The technology wouldn’t work. There’s too much variance in things like the angle the camera is viewing from versus glare from the windshield to get a reliable corpus of data to train from. Even then, the seatbelt isn’t super visible, it’s easily obscured by shadow or body shape or even the colour of your clothing.
The false positive rate would probably match the true positive rate - which is to say it’s a total coin toss and not something you want to rely on for enforcement
In fairness though, I think that’s probably the same for inattentive driving. You probably could do some video processing instead of simple image classification, but that’s a massive increase in complexity for dubious gain.
They only kill yourself if not used, and not others.