Picture published by the creator on GitHub.
Add boxes, tracking and zone counts on top of any vision model
Made by Roboflow
View profile on GitHub (opens in a new tab)Supervision is a Python library of the parts that sit around a computer vision model rather than inside it: loading datasets, drawing boxes and masks, tracking an object across frames, counting what crosses a zone, and measuring how well a model did.
It is written to be model-agnostic, so a classification, detection or segmentation model from anywhere — YOLO and many others — can be plugged in and get the same drawing, tracking and counting tools around it. Maintained by Roboflow, a computer-vision infrastructure company, it needs Python 3.10 or newer and is installed with pip, conda or mamba.
You could use it to…
- Count people crossing a doorway on camera
- Draw boxes and IDs on top of any detection model
- Track the same object as it moves across frames
Can I use this?
You'll need Python 3.10 or newer · Setup needed
Worth knowing It supplies the tooling around a model, not a model — you bring that. Counting people or vehicles from video raises privacy questions that a library cannot answer for you. MIT licence. BuildTube has not run or verified this software.
Why it's here
Added 25 September
51,041 GitHub stars as of Sept 25
- 51,041 people have starred it on GitHub.
Numbers checked on 25 September 2026; not refreshed since.
Behind it
See the code on GitHubBuildTube has not run or verified this project. Everything above is written from what the creator published.
Made this?