Frequently Asked Questions
The questions we hear most often on video analytics projects, with straight answers. If your question is not here, write to us.
General
What does CX Teknoloji do?
CX Teknoloji is a computer vision and AI company based at GTU Technopark in Gebze/Kocaeli, Türkiye. Using AI that runs on the streams of your existing IP/CCTV cameras, it delivers solutions for occupational health and safety, zone and perimeter security, production efficiency (OEE/MES), retail analytics and industrial machine vision; it offers 23 ready solutions across 5 categories.
Which industries do you serve?
Manufacturing and factories, warehousing and logistics, energy and hazardous-materials facilities, construction sites, retail stores, and quality control on production lines. Any site where a measurable event or metric can be extracted from camera footage is in scope.
What is EscherSense, and how does it relate to CX Teknoloji?
EscherSense is CX Teknoloji’s retail brand: it covers store analytics (people counting, demographics, heatmaps, queue management) and digital labelling solutions. It uses the same inference pipeline; its canonical address is cxteknoloji.com/eschersense/.
Where are you located, and which regions do you serve?
We are headquartered at GTU Technopark, Gebze / Kocaeli. We run projects across Türkiye and in Europe; we work in Turkish and English.
Technical
Does it work with our existing cameras?
Yes. The system runs on the RTSP stream of your existing IP and CCTV camera infrastructure; no additional hardware investment is required. We assess your cameras’ suitability free of charge before installation.
How many cameras can a single server handle?
A single GPU server typically processes 8–16 cameras in real time, depending on the solution type. For heavier models such as pose analysis, this drops to the 6–10 range.
What is your accuracy rate?
It depends on the solution and the site; in our reference projects we measured 96% vehicle safety accuracy and 96.2% counting accuracy. Accuracy is a field value, not a lab value: during the pilot, real events are labelled and compared with the system’s detections.
How do you reduce false alarms?
An event is never generated from a single frame: the condition must persist for a set number of frames and a time window, after which repeat suppression is applied. Thresholds are calibrated on your own site during the pilot.
Edge (on-premise) or cloud?
Both are possible. Processing can run on an on-premise GPU server — low latency, and the data stays on site — or in the cloud. On-premise processing is preferred in most production and security scenarios.
Which systems can it integrate with?
MES, ERP/SAP, WMS, SCADA/PLC, POS, CRM and occupational health and safety software via REST API or webhook; IoT alarm relays, sirens and mobile notification channels are also supported.
What are the minimum camera requirements?
Typically an existing IP/CCTV camera, 1080p resolution, and the monitored object appearing in the frame at a sufficient pixel height (e.g. a person’s body ≥ 80 pixels for PPE detection). Suitability is checked free of charge during the discovery phase.
Process and pricing
How long does installation take?
For ready solutions, installation and commissioning typically take a few weeks. For custom development, the timeline is set after the feasibility study; a typical custom project goes live within 12–14 weeks.
How does the pilot work?
The system is installed at the selected site using the existing cameras; polygons, thresholds and time windows are tuned on site, and false positives are reviewed daily. At the end of the pilot, the decision is made based on measured accuracy and event statistics.
How is pricing structured?
Pricing is project-based: the quote is prepared according to the number of cameras, the number of solutions, on-premise or cloud deployment, and the integration scope. The camera suitability assessment and the 15-minute demo call are free of charge.
How can I get a demo?
Reach us through our contact page or at +90 549 774 86 46 / [email protected]; in a 15-minute call we will identify which solution can be tested with your existing camera.
Data and compliance
Where is the data processed, and is it KVKK-compliant?
Processing can run on an on-premise GPU server or in the cloud. Data is processed anonymously, no facial recognition is performed and no biometric records are kept; the installation is configured to comply with KVKK, Türkiye’s data-protection law.
Do you use facial recognition?
No. Our solutions detect people, objects, poses and behaviour; there is no identification and no biometric recording. In retail analytics, demographic age/gender estimation is anonymous and limited to aggregate statistics.
How long are event records retained?
The retention period is set during installation according to the customer’s policy; event clips and metrics are stored on your own server or in the cloud of your choice.