RETAIL
Shelf & Planogram Compliance
Shelf & Planogram Compliance. What it takes to run it.
Detect out-of-stocks and incorrect product placement automatically.
// 01 · The question
“Is the right product in the right facing, and what is out of stock?”
// 02 · Why it is hard
The class count is enormous, hundreds of SKUs, many visually near-identical, and it changes constantly. Distinguishing two flavours of the same brand at shelf angle is the hard part.
// 03 · What the labels have to be
Bounding box
Decided before the first frame
Compliance is checked position by position against a planogram grid. Boxes map directly onto facings, and the comparison the system runs is between a grid of detections and a grid of expectations.
This is the decision that is expensive to reverse. The geometry has to match what the answer contains, and finding out it does not means labelling the set a second time.
// 04 · How we run it
Three parts of one loop, on this job.
LexAnnotate
Shopper Tracking
Annotate in-store video for customer detection, employee identification, door events, and product placement. Distinguish between customers, staff (by uniform), and children within each camera's region of interest.
LexInsight
Conversion Analytics
Turn raw foot traffic data into conversion intelligence. Link visual detections to store zones, time slots, and operational events. Surface patterns that static dashboards miss.
LexAlert
Store Alerts
Monitor detection accuracy across store locations. Detect when camera repositioning, seasonal displays, or lighting changes degrade model performance. Route the failures back for retraining.
// 05 · What breaks it after launch
Packaging refreshes are constant and the model is never told. The new pack is correct on shelf and unrecognised by the model, which reports an out-of-stock on a full facing.
It is not the only one that can get this use case, it is the one that usually gets it first. All ten conditions.
// 06 · In retail
Store Intelligence
The Challenge
The models lacked high-quality labeled data, activity alerts came back false, and unmonitored registers left customers waiting at checkout.
Our Execution
Labeled customer and staff positions. Validated alerts and traffic counts for accurate footfall tracking.
The Result
Faster AI training with accurate labeling. Smarter alerts for open doors and empty registers.
4M+
Annotations & validations