Camera and evidence layer
Video Core
Live monitoring, playback, camera health, event evidence, users, roles, audit logs and multi-site camera operations.
KVABI Architecture
From existing cameras to AI workflows, alerts and reports.
KVABI is an visual intelligence layer that connects to selected CCTV or IP camera streams, processes events on edge AI infrastructure and converts video into operational workflows for access, attendance, vehicles, safety, facilities, retail and service operations.
Works with existing cameras and VMS streams
KVABI connects to selected CCTV, IP camera, NVR or VMS streams already available at the site where technically supported.
Adds the VMS operations layer
KVABI VMS gives teams live monitoring, playback, camera health, evidence, user roles and audit history.
Runs close to the site
Edge AI devices process selected camera views locally for faster response and better control.
Creates operational workflows
Video events are converted into alerts, evidence, dashboards, reports and integrations.
Architecture principle
Video stays useful. Events become actionable.
KVABI keeps camera infrastructure in place and adds the workflow layer customers need to act on what cameras already see.
Platform layers
VMS, AI modules and integrations each have a clear job.
KVABI VMS should be understood as the video operations layer. KVABI AI modules sit above it to create intelligence and workflows. Integrations connect those events to site systems.
Visual intelligence layer
KVABI Intelligence
KVABI converts visual activity into decisions, alerts, actions, investigations and business records.
Connected action layer
KVABI Integrations
Integrations connect KVABI intelligence with barriers, turnstiles, access systems, dashboards, APIs, webhooks and operational systems.
How KVABI works
A simple flow from camera view to operational action.
Customers do not need to understand every technical component. The important point is that KVABI converts selected video views into structured operational events that teams can act on.
Step 1
Camera input
Selected CCTV, IP camera, RTSP, ONVIF, NVR or VMS streams are mapped to KVABI use cases where available.
Step 2
VMS operations layer
KVABI VMS manages live view, playback, camera health, evidence, users, roles and audit logs for the connected video operation.
Step 3
Edge AI processing
Local edge devices process relevant frames and run KVABI AI modules near the site.
Step 4
Module intelligence
Modules such as Sentinel, Presence, PlateVision, ParkGuard and OperatorAssist detect operational events.
Step 5
Workflow rules
Events are filtered through configured rules, zones, thresholds, timings and escalation logic.
Step 6
Alerts and evidence
Qualified events produce notifications, evidence snapshots, logs and review-ready records.
Step 7
Dashboards and reports
Operations teams get live visibility, analytics, history, reports and exportable evidence.
Reference architecture
The architecture is modular, edge-first and workflow-driven.
KVABI can be deployed as a focused module for one use case or as a multi-module AI layer across many operational zones and sites.
Deployment models
Choose the right deployment model for your site and governance needs.
KVABI can support local, hybrid and cloud-connected operating models depending on privacy, latency, connectivity and scale.
Fully local / on-premise
Best for high-security, offline-first or privacy-sensitive environments where video processing should stay within the site.
Local edge AI processing
Local database option
Works without constant internet
Useful for secure campuses and government deployments
Edge with central dashboard
Best for multi-site operations where each site processes locally but management needs centralized visibility.
Local site processing
Central reporting dashboard
Multi-location event visibility
Useful for warehouses, parking sites and communities
Cloud-connected workflow
Best for teams that want remote monitoring, centralized analytics, easy access and cross-site expansion.
Remote dashboard access
Cross-site reports
Centralized administration
Useful for enterprise and partner-led rollouts
Edge AI processing
Process locally. Send events, not unnecessary video.
KVABI is designed around practical site operations. The edge layer helps process selected video streams locally and convert them into lightweight, structured events for the dashboard.
Local inference
Camera streams are processed on edge devices so alerts can be created close to the source.
Offline-tolerant operation
Selected workflows can continue locally even when internet connectivity is unstable or unavailable.
Event-first storage
KVABI focuses on structured events, snapshots and logs instead of forcing continuous video uploads.
Low disruption deployment
Most deployments start by connecting to existing camera streams and configuring zones and rules.
Modular architecture
One platform. Multiple operational modules.
Start with one workflow, then expand to more modules, zones, cameras and sites when the operational value is proven.
SentinelPresenceGateAssistPlateVisionParkGuardOperatorAssistSafeGuardSpeedGuardElevatorAssistBayFlowRetail Intelligence
Integrations
Connect AI events to the systems that run the site.
KVABI can trigger or support workflows around access control, parking barriers, dashboards, local devices and enterprise integrations.
Access systems
Integrate with doors, turnstiles, QR workflows, visitor approvals and identity-based access events.
Vehicle gates and barriers
Connect vehicle recognition, pass validation, queue alerts and gate workflow decisions.
Dashboards and APIs
Send events, alerts, reports and operational data to external dashboards or enterprise systems.
RTSP / ONVIF and VMS streams
Connect selected RTSP, ONVIF, NVR or VMS stream sources where available and technically supported by the site design.
Printers and local devices
Support workflows such as parking receipts, visitor slips, gate passes and local operational output.
Security and governance
Designed for practical enterprise governance.
KVABI architecture supports privacy-aware deployment, local processing options, event-based records and role-based access for operational teams.
Local processing control
Sensitive camera analytics can be processed locally based on site and governance requirements.
Role-based access
Dashboards, event history and reports can be limited to authorized users and operational roles.
Evidence governance
Event snapshots, logs and reports can be controlled based on retention and review policies.
Integration boundaries
Barrier, turnstile, printer, API and third-party integrations can be configured with clear control logic.
Deployment roadmap
A clear path from first camera to scaled operational AI.
1
Site discovery
Understand the operational problem, camera availability, site zones, user roles and success criteria.
2
Camera and zone mapping
Map selected camera streams to gates, lanes, counters, bays, aisles, doors, queues or safety zones.
3
Edge deployment
Deploy the local edge AI device, connect streams and configure module-specific detection workflows.
4
Workflow configuration
Define thresholds, alerts, evidence snapshots, escalation rules, reports and integration requirements.
5
Pilot validation
Validate event quality, false positives, missed events, uptime, user workflow and operational value.
6
Scale and optimize
Expand across more cameras, modules, zones, sites and integrations based on validated outcomes.
Buyer questions
Architecture questions customers usually ask first.
This section helps buyers quickly understand deployment requirements before a detailed technical discussion.
Do we need to replace existing cameras?
Usually no. KVABI is designed to work with selected existing camera streams where angle, lighting and coverage are suitable.
Does video need to go to the cloud?
Not necessarily. KVABI can run with local edge processing and local deployment models where cloud connectivity is not required.
Can it work across multiple sites?
Yes. Each site can process locally while central teams view alerts, reports and dashboards across locations.
Can it connect to barriers, turnstiles and other systems?
Yes. KVABI can support workflow integrations with barriers, turnstiles, access systems, printers and APIs depending on the site design.
How is it different from a VMS?
KVABI VMS provides the video operations layer for live view, playback, camera health, evidence, users and audit history. KVABI AI modules then turn selected video feeds into operational workflows, alerts, evidence and reports.
Want to map KVABI architecture for your site?
Share your site type, camera availability, operational use case and integration needs. We can help recommend the right edge, module and dashboard architecture.