The retail shrinkage problem and why traditional methods fail at scale
By 2024, shrinkage costs across global retail exceeded $100 billion each year, according to the NRF National Retail Security Survey—a published estimate rather than an operational benchmark. Traditional loss prevention — EAS tags, human observation, receipt checks — catches a fraction of losses and scales poorly. Adding more security staff is linearly expensive; adding more cameras without intelligence just creates more footage that nobody watches.
Computer vision for loss prevention promises automated detection: identifying concealment events, scan-avoidance at self-checkout, ticket switching, and organised retail crime patterns. The promise is real — deployed correctly, CV-based LP systems detect events that human operators consistently miss. But the gap between demonstration accuracy and production reliability is where most deployments struggle.
Why is loss prevention CV harder than general object detection?
Thousands of SKU variants under fluctuating store lighting challenge CV-based loss prevention systems, causing off-the-shelf models to degrade rapidly at scale. The specific challenges:
SKU diversity. A typical grocery store carries 30,000–50,000 SKUs. Self-checkout fraud detection requires distinguishing between items that look similar (varieties of the same product, similar packaging across brands). Generic object detection models trained on COCO or ImageNet do not have this granularity.
Lighting variability. Retail environments have mixed lighting — fluorescent overheads, natural light from windows that changes hourly, seasonal variation, refrigerator lighting. Models trained in controlled conditions degrade when lighting shifts outside the training distribution.
Occlusion and angles. Shoppers’ bodies, carts, bags, and other products occlude the items of interest. Overhead cameras capture a fundamentally different view than aisle-level cameras. Multi-angle systems are necessary but multiply the pipeline complexity.
Normal vs suspicious behaviour. Customers routinely handle, examine, and put back products. The difference between “examining an item” and “concealing an item” can be a matter of milliseconds and millimetres in hand position. False-positive rates on concealment detection are typically 5–20× higher than on simple object detection (an observed pattern across our retail CV engagements, not a published benchmark).
The compound detection pipeline
Detection, tracking, and POS reconciliation must work together; relying on a single model floods teams with unactionable alerts. A production LP pipeline requires:
| Stage | Function | Technology |
|---|---|---|
| Person detection & tracking | Maintain identity across camera views | Multi-object tracker (ByteTrack, DeepSORT) |
| Item detection | Identify products being handled | Domain-trained detector (YOLO + SKU-specific fine-tuning) |
| Action classification | Distinguish normal handling from concealment/skip | Temporal action model (SlowFast, VideoMAE) |
| POS reconciliation | Match scanned items against detected items at checkout | Event correlation engine |
| Alert filtering | Suppress false positives using contextual rules | Rule layer with per-zone thresholds |
The POS reconciliation stage is what transforms noisy video detection into actionable intelligence. A concealment detection alone has limited value — the same event correlated with a subsequent checkout where the item does not appear in the scanned list becomes an actionable loss event. In our experience, this correlation layer is the single biggest determinant of whether the LP team trusts the system enough to keep it running.
What “works” looks like in production
Zero false positives remain unattainable in production-grade retail LP deployments. They achieve a false-positive rate that is low enough for the LP team to investigate every alert — an observed range below roughly 50% false-positive rate for high-confidence alerts, across our deployments rather than a benchmarked threshold. This is achieved through:
- Zone-specific models trained on each store’s camera geometry and lighting
- Confidence cascade — only alerts above multiple threshold stages reach human review
- Temporal confirmation — a single-frame detection is never an alert; sustained detection across frames is required
- POS correlation — alerts without corresponding POS anomalies are suppressed
The ROI that computer vision delivers in retail depends on this pipeline maturity. Immature deployments (single-model, no POS integration, no zone calibration) generate alert fatigue and negative ROI. Mature deployments reduce shrinkage by measurable percentages — but typically require 3–6 months of on-site calibration to reach that maturity (an observed planning heuristic across our retail engagements, not a vendor guarantee).
Scale is the hard problem
Scaling from a pilot—10 cameras in one location—to 5,000 cameras spanning 500 stores introduces failure modes invisible in small tests. Scale introduces: model drift across geographically diverse lighting conditions, fleet management for edge inference hardware (TensorRT-optimised models on NVIDIA Jetson or equivalent), centralised alert triage across hundreds of locations, and the statistical certainty that even a 1% false-positive rate (as an illustrative threshold) generates thousands of daily false alerts across the fleet.
Loss prevention CV at scale is an infrastructure and operations problem as much as it is a machine learning problem. Teams that treat it as only a model accuracy challenge discover the operations gap after deployment — when the alert volume overwhelms the LP team and the system is disabled store by store. We pay close attention to this transition point when scoping retail CV engagements, because the right answer at 10 stores is almost never the right answer at 500. We run that transition-scoping work under the same Computer Vision R&D practice that covers our retail loss-prevention engagements. That scoping work is what our retail computer-vision practice runs before a pilot is allowed to become a fleet-wide commitment.
