Detection on the Vivotek Cameras Already on Site
No Vivotek-specific connector is prebuilt. The connection would be to the stream, on the standard RTSP and ONVIF path that already carries Axis, Hanwha and Verkada, so an engagement scopes configuration and verification rather than development. Vivotek is an IP camera manufacturer used in retail, transport and municipal installations, and the cameras would stay exactly where they are. AI Live Insight would read the feeds.
How it connects
Each camera would be added by pointing the portal at its existing feed: domain or IP, port, stream name, and a username and password where the camera requires them. ONVIF profiles S and T cover discovery, stream URL retrieval and PTZ, which is the part of ONVIF that is universal across the major vendors. Encoding would stay on the camera, the existing H.264 or H.265 bitstream read as emitted with the codec detected per source, so nothing is re-encoded and no encoder load is added.
Each camera would stand on its own with its own connection details, detection settings and recordings, and appear in the library as a tile that can be watched, organized into folders, searched and sorted. Detections would become a timed metadata track beside each recorded segment, and the clip around an event would be filed under one Recording Folder organized by camera, under the library's access control, retention and chain of custody.
The connection is to the stream, so camera control and device management would stay on the Vivotek side. A camera's own on-device analytics would not be consumed either: platform ONVIF support today is profiles S and T, and the metadata and event profile is not part of it, so detection would run on the stream instead.
What you can do together
- Add detection to Vivotek cameras already installed, with no camera replaced and no hardware added beyond the server that runs the analysis.
- Group cameras into folders that match the site, which is worth doing beyond a handful.
- Keep the clip around an event as evidence under retention and a custody record.
- Configure classes, thresholds and zone rules per camera rather than once for the estate.
A scenario
- ScopingThe engagement would confirm which cameras cover the exits, how each is reached on the store network, and what counts as blocked rather than briefly passed.
- SetupA server would be placed centrally or per region depending on the network, and the stockroom cameras added and grouped into a folder per store.
- ConfigurationA zone rule would cover the exit route, with a dwell condition so a trolley passing through does not alert.
- A SaturdayA pallet left in the route would raise an alert, with the clip filed in the library under that store's folder.
- ReportingThe safety manager would pull the quarter's clips by store, and the stores with repeat findings would be evident from the record.
What stays where
Vivotek keeps the cameras
Device settings, firmware and camera control stay on the Vivotek side.
The stream is the whole of the connection
Inbound only, with nothing written back to any camera.
On-device analytics stay on the device
Platform ONVIF support is profiles S and T, so a camera's own analytics events are not consumed.
Processing would run on site
A server on the same network as the cameras is what keeps alert latency low, and it is the deployment this work is projected for.
Products and solutions
Next step
See it on your own Vivotek instance. We will show the connection made, the data moving and the output, then size it for your deployment.
Request a demonstration or write to sales@vidizmo.ai
sales@vidizmo.ai · vidizmo.ai/integrations/catalog/vivotek