CX Teknoloji
TR·EN·AR
ÜRT-04 · Production & Operations

Custom Video Analytics

Models built from scratch for needs that off-the-shelf solutions cannot meet.

12–14 weeks
Typical time to go live
4 stages
Feasibility → training → test → live
100%
Site-specific training data

Every site has a problem of its own: a defect type that exists only on that line, a movement that means something only in that facility.

For such needs we start with feasibility: we state plainly what can be measured with the existing footage, what accuracy is realistic and what the labelling effort costs. Data collection, labelling, model training, field testing and go-live follow — a typical custom project goes live within 12–14 weeks.

How it works

01 / FEASIBILITY

Is it measurable?

Samples of your existing footage show whether the target is technically achievable.

02 / DATA

Collection and labelling

Representative data is collected on site; rare cases are sought out specifically.

03 / GO-LIVE

Test and calibration

Accuracy is measured during the pilot and thresholds tuned to the site.

Accuracy is a target; how it is reached is a commitment

In custom projects the crucial point is defining up front how accuracy is measured. What counts as a “correct detection”, how many false alarms are acceptable and which dataset the test runs on are written into the contract. The curve below shows the path a typical custom model follows during a pilot.

Share of project duration by stage · 12–14 week project
Feasibility and scoping2 weeks
Data collection and labelling4 weeks
Model training3 weeks
Field test and calibration3 weeks
Go-live and user training1 week

Data, not the model, takes the largest share of the timeline; capturing rare cases determines project quality.

Accuracy through the pilot · 71% → 94%
W171
W276
W382
W486
W589
W692
W793
W894

The curve rises as each week’s false detections from the site are added to the training data.

Use cases

Quality

Surface defect

Detection of line-specific defect types.

Agriculture

Produce grading

Separation by size and colour.

Transit

Passenger behaviour

Station-specific flow analysis.

Energy

Equipment state

Reading visual gauges and flagging signs of faults.

Technical requirements

StartFeasibility on 1–2 weeks of sample footage from your site
DataLabelled by our team, with your approval
HardwareCustom camera/lighting design included where needed
DeliveryModel, panel, API and user training
MaintenancePeriodic retraining and accuracy reporting

Frequently asked questions

How does Custom Video Analytics work?

Every site has a problem of its own: a defect type that exists only on that line, a movement that means something only in that facility.

What results and indicators does it provide?

Key indicators: Typical time to go live: 12–14 weeks; Feasibility → training → test → live: 4 stages; Site-specific training data: 100%. Data, not the model, takes the largest share of the timeline; capturing rare cases determines project quality. Thresholds are calibrated on your own site during the pilot.

Is it compliant with KVKK, Türkiye's data-protection law?

Footage is processed on an edge server inside your facility; raw video is not sent to the cloud. Facial recognition is not used; analytics outputs are anonymous and event clips are deleted automatically when the defined retention period ends.

See this solution on your own site
Let's test it with your existing camera in a 15-minute demo call.
Book a 15-min demo

Other solutions in Production & Operations

ÜRT-01 · OEE
Production Efficiency & OEE
Measure downtime, cycle time and bottlenecks from video.
ÜRT-02 · Product ID
Product-Based Video Evidence
Match and archive every package’s footage with its product ID.
ÜRT-03 · 5S
Housekeeping & 5S Monitoring
Automate checks for clear walkways, spills and unattended objects.

Technical articles on this topic (in Turkish)

RTSP'den Edge GPU'ya
Görsel-Dil Modelleri
Yeni Nesil Edge AI