Customers & Case Studies
The case studies below are real field results that we also publish on our home page. Every figure is an accuracy or safety indicator measured on the customer’s own site during the pilot; we don’t invent numbers, and we don’t write about what we haven’t measured.
City Lojistik
Where forklift and pedestrian traffic crossed in the warehouse aisles, risk went unnoticed in the moment; and when a shipment was disputed, there was no way to prove the condition the product had been loaded in.
Vehicle–pedestrian proximity detection and zone rules on the existing cameras; product-based video evidence recording in the shipping area.
Emilio
How many people entered the store, which areas they stopped in and how much of that turned into sales were all managed by guesswork.
People counting, zone-level dwell time (heatmap) and conversion measurement with the existing cameras — no additional sensors, anonymous and GDPR/KVKK compliant.
Textile manufacturer
Nobody measured at which station time was lost, or whether it came from waiting or walking, so line balancing relied on estimates.
Existing cameras measured active work, waiting and walking time per station. Reports assess the station, not the person; no face recognition, data stays anonymous. Once waiting losses were visible, the line was rebalanced; the data feeds the MES.
How we measure results
On site, with your data
Accuracy is a field value, not a lab value: throughout the pilot, real events are labelled by hand and compared with the system’s detections.
Suppression with a time window
An event is never generated from a single frame; the condition has to persist for a set number of frames and a set duration. False positives are reviewed daily during the pilot.
A clip for every event
Every detection is archived as a timestamped clip and a dashboard metric, and serves as evidence in disputes, audits and training.
Named references and case studies are published only with the customer’s written permission; other cases are shared anonymously.