Customer counting: Datasets & Models
Automatically count customers in retail spaces using object detection on video or still images from existing cameras. Track how many people enter, browse, and move through defined zones without manual tallying.
Customer counting
This customer counting workflow applies object detection to identify and count individual people in retail spaces, using video streams or still images from existing store cameras as input. It detects each person as they enter, browse, or move through defined zones, producing running headcounts that support customer counting detection without manual tallying.
Try it out
Powered by general detection model
Run on custom image
Drop an image here or click to upload
Why automate customer counting
Manual customer counting with clickers, door sensors, or staff estimates is labor-intensive, error-prone, and impossible to scale across multiple locations. A vision-based approach turns any existing camera into a reliable counting instrument, capturing entries, exits, and zone traffic continuously without adding hardware or staffing. For retail tech teams, this means grounding decisions about staffing, store layout, promotions, and conversion rates in real occupancy data rather than guesswork.
Key benefits
Customer counting detection runs on cameras you already have, so deployment does not require costly new infrastructure. It scales cleanly from a single entrance to a network of stores, delivering consistent, timestamped headcounts that feed dashboards, staffing models, and conversion analysis. Using a model like RF-DETR, you get strong detection accuracy in crowded and variable lighting conditions typical of retail floors.
How to use it
- Upload an image: Drop any photo into the widget above to try customer counting for free, right on this page.
- Preview instant results: The model detects person, shopping cart, shopping bag, stroller 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 customer counting
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 customer counting, fine-tune it on your own labeled images, then deploy to a Serverless Cloud API endpoint.
Not sure where to start?
Describe your Customer Counting project to Roboflow Agent.