Hi; not responding to the question, since by definition I cant provide Independent testing. I would like to take a second to explain what OZT is and does.
OZT stands for Observe/Zoom/Track. We use two standalone cameras - one of our newer model Hemispheric cameras, and a PTZ or Speed Dome (at this time, auto tracking capable models only).
The Hemispheric camera will "command" the PTZ/Speed Dome to look at a position where an event has occurred (Video Motion Detection right now, more analytics later), at which point the Dome will track the object that created the event, until the object leaves it's view, or another event occurs in the Hemispheric cameras field of view.
Happy to address any questions, but again, just wanted to give a quick blurb for anyone that didn't want to watch the video clip that John linked to 8-)
I have seen similar setup from other brand, but what will happen when multiple subject appeared in the panoramic field of view?
Will the built in video analytic algorithm have to be set to a certain parameter right? assuming the adjustment will be base on subjects' size, speed and the given clarity/noise free condition off course.
What about other parameter settings like line crossing, directional, object appearance/disappearance? We all know that the VCA is somewhat always have some margin of error, how does this better in terms of performance compared to the other independently run VCA by a separate server?