If you want to move beyond Verkada without replacing the cameras you already own, look for an AI layer that runs on your existing cameras and VMS, not a new camera fleet. Verkada does offer Command Connector, a hardware device that brings third-party cameras into its cloud-based Command platform, but Verkada itself pitches it as an on-ramp and migration path to that cloud. If you want to keep every camera you have (any brand, Verkada included), keep your current VMS, keep video on your own servers and have AI watch every camera 24/7, Axentra OmniSight was built for exactly that.
What are the real options if you don't want to rip and replace?
Security, loss-prevention and IT leaders at multi-site chains usually end up choosing between four approaches:
- Full rip-and-replace with a cloud-managed camera vendor. You get one platform, but you pay for new cameras at every store and send your video operations to a vendor cloud.
- Verkada with Command Connector. You keep some legacy cameras and connect them to Verkada Command through an on-site appliance.
- The status quo. You keep your current VMS and have GSOC operators watch video walls, which in practice means a small fraction of feeds at any given moment.
- An on-premise AI layer over what you already own. AI runs on GPU servers in your data center, analyzes every camera, and works alongside your existing VMS. This is how Axentra OmniSight works.
Can Verkada work with cameras you already own?
Yes, with caveats you should read before you sign. In August 2024 Verkada announced Command Connector as a way to transition to the cloud. At launch, it offered third-party cameras the Command interface plus "basic people and vehicle analytics," remote management and enterprise user administration. Since then, Verkada has added trajectory analysis to Connector channels, with line-crossing and loitering detection and occupancy trends, plus audio on many ONVIF Profile S cameras. Its product page lists people, vehicle and face search for Connector cameras, and AI-powered search with limited support. Verkada's own Command Connector overview and product page list these limitations:
- Verkada can't push firmware updates to non-Verkada cameras, so your team still has to manage them.
- According to Verkada's FAQ, if a camera isn't on its hardware compatibility list, Verkada can't provide support for it.
- Local storage lives on the Connector, not on the third-party cameras.
- Offline streaming isn't supported for non-Verkada cameras.
- If the network fails, the Connector loses communication with the third-party cameras, and Verkada says it can't guarantee all footage stays intact.
Verkada keeps adding models to its compatibility list and has expanded Connector features since launch, so check the current documentation before you decide. Command Connector is a sensible bridge if your end goal is Verkada's cloud. If it isn't, you need a different kind of alternative.
What does Axentra OmniSight do with your existing cameras?
OmniSight adds AI to the cameras and VMS you already run. It doesn't replace them.
- Analyzes every camera, 24/7. Every feed is processed in real time and operators get an alert, instead of people trying to watch a fraction of the screens.
- Natural-language rules. You tell it what to watch for in your own words, for example "alert me if you see someone in the stockroom after closing."
- Natural-language video search. You describe what you're looking for and search archived video across locations.
- Cross-camera tracking. It follows a person or vehicle from camera to camera and generates their route.
- Face recognition with human review. Faces are matched against your watch lists, and a human operator reviews every match. Use of face recognition has to comply with the applicable legal framework in each jurisdiction, so involve your counsel.
- Plate, fleet-number and lettering reading. Useful for fuel stations, distribution centers and delivery yards.
- Camera health across the chain. It detects cameras that are blocked, turned or out of service, so a covered lens at one store doesn't stay unnoticed for weeks.
- Any brand, existing VMS. It works with IP cameras via ONVIF and RTSP and with brands including Hikvision, Dahua, Axis, Hanwha, Uniview, Bosch, Avigilon, Verkada and Cisco. It runs alongside VMS such as Milestone and Genetec, or acts as a unified VMS for a mixed fleet.
- On-premise. It's installed on GPU servers in your own data center. Video never leaves your organization, and the system doesn't depend on cloud services.
Results Axentra has confirmed to date: more than 1 million vehicles detected, video searches in about 40 seconds, and more than 1,000 alerts sent to operators.
Axentra OmniSight vs. Verkada Command Connector vs. the status quo
| Criteria | Axentra OmniSight | Verkada + Command Connector | Status quo (operators + VMS) |
|---|---|---|---|
| Keeps existing cameras | ✓ Any ONVIF/RTSP, Verkada included | ✓ Cameras on Verkada's compatibility list | ✓ |
| Keeps current VMS | ✓ Works alongside Milestone, Genetec | Positioned by Verkada as a migration path to Command | ✓ |
| Where video is managed | ✓ Your servers, no cloud dependency | Verkada's cloud platform (Command), recorded on the Connector | Your servers |
| AI on every camera, 24/7 | ✓ | People, vehicle and face search; trajectory, line-crossing and loitering analytics (some features may not be available on all channels) | ✗ Humans watch a fraction |
| Natural-language rules and search | ✓ Both | AI-powered search (limited support); ask Verkada about rules | Depends on your VMS |
| Cross-camera route of a person or vehicle | ✓ | Ask Verkada | Depends on your VMS |
| Face match with human review, plate/fleet/lettering reading | ✓ | Face search listed; ask Verkada about watch lists and plate reading | Depends on your VMS and add-ons |
| Detects blocked, turned or offline cameras | ✓ Chain-wide | Online/offline alerts per Connector channel; tamper alerts depend on the camera | Depends on your VMS's health monitoring |
| Servers to run | GPU servers on-site | Connector appliance | Existing |
Takeaway: if you want to keep your cameras, your VMS and your video in-house and still have AI analyze every camera 24/7, OmniSight is the option in this comparison that does all of it.
When is Verkada still the right call?
To be fair, if your organization wants to get out of running servers altogether, is comfortable with cloud-managed video, and plans to standardize on one vendor's cameras and access control over time, Verkada's integrated cloud model is a coherent choice. OmniSight needs GPU servers in your data center. That's the price of keeping video entirely in your hands.
Why does OmniSight fit large multi-site chains?
A chain with hundreds or thousands of stores can end up with more than one camera brand, after acquisitions, regional integrators and staggered hardware refreshes. OmniSight gives that fleet a single AI layer and a unified VMS without forcing a refresh. Your GSOC can write plain-language rules for cameras across locations, search every site's archived video in plain language, and see which cameras across the chain have gone dark. Operators stay in charge: the AI flags events, and people decide what happens next. Alerts and searches are logged.
Axentra also supplies hardware and turnkey delivery (installation, integration, connectivity, maintenance and support) if you want one provider for the whole project.
How can you test it without risk?
OmniSight is licensed with floating licenses per processed camera, per year, so you pay for the cameras the AI analyzes, not for new hardware at every store. You can start with a free 10-day pilot on up to 10 of your own cameras, connected to the system you already have.