VisionAI
Give your existing CCTV cameras AI eyes — footfall and demographics out front, food safety and behaviour monitoring out back.
Definition
Codemax VisionAI is a computer-vision platform that empowers existing CCTV cameras with AI capabilities — front-of-house demographic and footfall monitoring, and back-of-house food-safety and employee-behaviour monitoring — without requiring new hardware.
01 · Overview
What is VisionAI?
VisionAI runs computer-vision models against the RTSP or ONVIF streams your existing cameras already produce. Front-of-house, the cameras that cover your floor become a continuous source of footfall, demographic, and dwell data. Back-of-house, the same estate watches food-safety compliance — PPE, hand-washing, station hygiene — and flags unsafe employee behaviour in real time. Inference can run on the edge, on-prem, or in the cloud depending on data-sensitivity requirements.
02 · Audience
Who is it for?
- F&B operators enforcing kitchen food-safety and PPE compliance
- Retailers measuring footfall, dwell, and demographics
- Central kitchens monitoring hygiene-critical production zones
- Healthcare, MICE, and venue operators tracking safety and crowd flow
03 · Capabilities
Key features
Every feature ships as part of the platform — no add-on SKUs, no surprise modules.
Front-of-house footfall and demographics
Anonymous footfall, age-band, gender, and dwell analytics at any monitored zone — no PII captured.
Back-of-house food-safety monitoring
PPE, hand-wash, uniform, and station-hygiene detection with shift-level and station-level reporting.
Employee-behaviour monitoring
Detect unsafe or non-compliant behaviour in production and service areas — surfaced as coaching data, not surveillance.
Queue and service analytics
Queue length, wait time, and service-speed signals piped into operations dashboards.
Works with existing CCTV
Reads RTSP / ONVIF streams from any modern IP camera. No camera-vendor lock-in, no rip-and-replace.
Edge or on-prem deployment
Run inference on the edge or behind your firewall for data-sensitive environments.
See VisionAI running on data like yours.
Book a 30-min demo04 · Engagement
How it works
From scoping to network-wide operation, in clearly defined phases.
-
Survey your camera network
We audit existing camera placements and confirm coverage for the use cases you want to enable.
-
Connect streams
VisionAI connects to your cameras via RTSP / ONVIF — no rip-and-replace required.
-
Calibrate models
Models are tuned to your environment, lighting, and policies before going into production.
-
Operate with alerts and dashboards
Live dashboards and routed alerts go to operations, food-safety, and security teams — and stream into the AI-Kitchen Command Center.
05 · Pricing
Pricing
Custom pricing per camera, per use case, and per deployment topology (cloud, on-prem, or edge). Contact us — we will quote against your camera count and shortlisted use cases.
06 · Fit
Where it fits, what you get
Sectors served
- F&B
- Retail
- Healthcare
- MICE
Outcomes you can measure
- Continuous food-safety and behaviour compliance
- Footfall and demographic visibility without new hardware
- Faster incident response with real-time alerts
07 · FAQs
Frequently asked questions
The questions enterprise buyers ask before they sign — answered up front.
- Do we need to replace our existing CCTV?
- No. VisionAI works with any modern IP camera that streams RTSP or ONVIF. We have not yet found an estate we needed to replace.
- What does VisionAI monitor front-of-house vs back-of-house?
- Front-of-house: anonymous footfall, demographics (age band, gender), dwell, and queue analytics. Back-of-house: food-safety compliance — PPE, hand-washing, uniform, station hygiene — and unsafe employee behaviour. Each camera is assigned only the use cases you enable.
- Does VisionAI capture personal data?
- VisionAI is designed to operate on anonymised signals — age band, gender, dwell, PPE compliance, queue length. We do not perform face recognition or store identifiable biometrics by default. Identifiable use cases require explicit configuration and consent.
- Is VisionAI compliant with PDPA and GDPR?
- Yes. We deploy with PDPA (Malaysia) and GDPR (EU) in mind, including edge / on-prem options for data-residency requirements and configurable retention windows.
- Can VisionAI run without sending video to the cloud?
- Yes. Edge and on-prem deployments are supported. Only event metadata (e.g. "PPE violation at Camera 3, 12:04") needs to leave your network if you choose.
- How accurate are the models?
- Accuracy depends on camera placement, lighting, and the use case. We benchmark on your footage during calibration and only go live once accuracy meets the threshold we agree with you.
- How does VisionAI integrate with the rest of Codemax?
- VisionAI events stream into the AI-Kitchen Command Center for real-time anomaly detection and dashboards. Operational triggers can also write back to RMS — for example, opening a food-safety incident in the operations log.
08 · The stack
Works better together
VisionAI shares one data spine with the rest of the Codemax platform — every solution you add compounds the others.