Person detection: Datasets & Models
Person detection locates and identifies people in images and video frames, drawing bounding boxes around each individual for downstream tracking, counting, or alerting. It's a foundational task for security and surveillance systems that need to know when and where people appear.
Person detection
This detection workflow locates people in images and video frames, returning a bounding box for each person it finds so you can feed the results into tracking, counting, or alerting logic. It accepts inputs from IP cameras, CCTV feeds, drone footage, or uploaded files, and models trained on a person dataset can handle crowded, low-light, or partially occluded scenes.
Try it out
Powered by general detection model
Run on custom image
Drop an image here or click to upload
Why automate person detection
Manually watching camera feeds for human presence is tedious, error-prone, and does not scale past a handful of monitors, and security staff inevitably miss events during long shifts. Automating person detection lets a system flag intrusions, count occupants, or trigger alerts the moment someone enters a restricted zone, without relying on constant human attention. For security and surveillance teams, this means faster response to incidents, reliable after-hours coverage, and searchable records of when people were present in a given area. It also frees operators to act on alerts rather than stare at screens.
Key benefits
You get consistent, round-the-clock detection that integrates directly with existing IP cameras and NVR infrastructure, so there's no need to replace hardware. The workflow scales from a single entrance camera to hundreds of feeds across multiple sites, and RF-DETR delivers strong accuracy on small and occluded subjects typical of surveillance angles. Detections can be piped into alerting, analytics, and access-control systems to turn raw footage into actionable events.
How to use it
- Upload an image: Drop any photo into the widget above to try person detection for free, right on this page.
- Preview instant results: The model detects person, face, backpack, helmet 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 person 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 person detection, fine-tune it on your own labeled images, then deploy to a Serverless Cloud API endpoint.
Not sure where to start?
Describe your Person Detection project to Roboflow Agent.