Shrinkage detection: Datasets & Models
A computer vision workflow that monitors retail environments to detect behaviors and events associated with inventory shrinkage, such as concealment, unscanned items, and unauthorized removal of goods. It runs object detection on camera feeds to flag potential loss events in real time.
Shrinkage detection
This workflow runs object detection on retail camera feeds to support retail theft detection by flagging events tied to inventory shrinkage. It identifies signals such as merchandise concealed in bags or clothing, unscanned items at checkout, and goods leaving through exits, taking still images or video frames from in-store cameras as input.
Try it out
Powered by general detection model
Run on custom image
Drop an image here or click to upload
Why automate shrinkage detection
Manual retail theft detection depends on loss prevention staff watching banks of monitors, which does not scale across dozens of camera feeds and misses events during busy periods. Shrinkage silently erodes margins in a low-margin retail business, and after-the-fact review of footage is slow and reactive. Automating detection with computer vision surfaces suspicious activity as it happens, letting a small loss prevention team focus attention where it matters. It turns passive recorded footage into an active signal you can route to alerts, dashboards, or point-of-sale exception reports.
Key benefits
Teams get consistent, around-the-clock monitoring across every camera without adding headcount, plus a searchable log of flagged shrinkage events for investigation and trend analysis. The workflow integrates with existing IP cameras and self-checkout hardware, so there is no need to rip out current infrastructure. Because it runs on RF-DETR, it scales from a single store to a multi-location fleet while maintaining detection accuracy in cluttered retail scenes.
How to use it
- Upload an image: Drop any photo into the widget above to try shrinkage detection for free, right on this page.
- Preview instant results: The model detects person, shopping bag, merchandise item, shopping cart, backpack within seconds and returns an annotated image with structured predictions, so you can immediately see how well it works. Adjust the detection classes and re-run until the results fit your scenario.
- Like what you see?: With a free Roboflow account, you can deploy this workflow with the Serverless Cloud API, customize it in your workspace, or train RF-DETR on your own images for production accuracy.
- Get started for free: No code or training data required to begin: upload, review the results, and make it part of your stack in a single click.
Recommended model for shrinkage detection
RF-DETR is state of the art for real-time object detection, instance segmentation, and keypoint detection, with higher accuracy and lower latency than comparable models, and it's Apache 2.0 licensed. When you're ready to build a production model for shrinkage detection, fine-tune it on your own labeled images, then deploy to a Serverless Cloud API endpoint.
Not sure where to start?
Describe your Shrinkage Detection project to Roboflow Agent.