Model A
Private On-site (edge GPU)
Video analysis runs on an edge GPU inside your plant. Inference media stays on site.
Best fit: production lines, restricted zones and sites where footage must remain on premises.
Wyse Envision · Industrial visual operations evidence
Factory visual checks that produce evidence, not just footage
Between shifts, nobody watches every camera. Manual rounds are inconsistent; incidents surface late; audits take days of footage review. Your CCTV records everything and proves nothing by itself.
L2 decision support: humans approve every consequential action. Accuracy is measured during a paid pilot on YOUR cameras — no generic promises.
One site, one workflow, a clear go / no-go. The Visual Operations Evidence Brief is sent when you book the clinic.
What Wyse Envision does
Scheduled visual checks on the cameras you already have
Wyse Envision runs scheduled visual checks on the cameras you already have: station presence/duration, restricted-zone exceptions, blocked exits, shift condition and housekeeping. Exceptions arrive with video evidence for a named reviewer — L2 decision support, human-approved actions.
- Work presence and duration at a station
- Restricted-zone exceptions and blocked exits
- Shift condition and housekeeping checks
- SOP verification with human review
- Incident evidence packs for safety and operations reviews
Deployment
Where the AI runs is your decision
Model A edge GPU: media stays on site (preferred for factories). Model B governed cloud: for sites without local hardware, only approved frames leave, under explicit data-transfer approval.
Model B
Managed Cloud (governed)
For sites without local hardware, only approved frames leave the site, under explicit data-transfer approval, with provider and region disclosed.
Best fit: trials and smaller sites without a server room. Offered as a controlled beta.
Proof before scale
Paid Site & Scenario Readiness Study, then a measured pilot
Paid Site & Scenario Readiness Study (RM3k/month × up to 3 months): camera survey, N/L/A sizing, workflow baseline, privacy map and written go/no-go. Then a paid pilot measured against agreed test cases — accuracy numbers you can trust because they are yours.
- Readiness study — camera census, sizing, scenario baseline, privacy map, written go / no-go.
- Paid pilot — one primary scenario on real cameras, acceptance criteria signed before it starts.
- Production — only after measured results. See how a pilot is accepted →
FAQ
Questions plant and EHS leaders ask first
Wyse Envision works with suitable RTSP/IP camera streams after site validation. Camera coverage, angles, lighting and network conditions are checked in the readiness study before any commitment.
Model A (edge GPU) keeps inference media on site. Model B (governed cloud) sends only approved frames off site under explicit, auditable data-transfer approval, with provider and region disclosed.
No. Biometric processing is excluded from the standard offer. Any biometric scenario would require its own legal framework and separate approval.
None is quoted generically. Accuracy is measured during your pilot against test cases agreed in advance, on your cameras.
Wyse Envision works alongside your existing NVR or VMS using tested generic interfaces. Native connectors are scoped as project deliverables after the readiness study.
Get started
Book the Industrial Site-Readiness Clinic
Bring the one plant you suspect most. In 30 minutes we look at the camera conditions and two or three candidate checks with you, then send a Readiness Study proposal within 48 hours.
See the Work. Use the cameras you already have to produce evidence of one operational condition that supervisors currently check manually.