Litter detection: Datasets & Models

Manually surveying streets for litter is slow, inconsistent, and hard to scale across an entire city, and complaint-driven cleanup often means trash sits for days before anyone reports it. Automating litter detection with computer vision lets Smart Cities and traffic agencies continuously monitor public spaces and quantify where waste accumulates, turning subjective inspections into measurable data. Instead of dispatching crews on fixed routes, teams can target the areas that actually need attention. This makes it possible to catch illegal dumping and overflow conditions early, before they escalate into larger sanitation problems.

Automated litter detection reduces the labor spent on visual street audits and produces objective, repeatable data on where and how much waste is present. It integrates with existing municipal camera infrastructure, dashcams, and drone programs, so cities can deploy it without buying new hardware. Because the model runs on any image feed, coverage scales from a single intersection to a citywide network as needs grow.

1. Upload an image: Drop any photo into the widget above to try litter detection for free, right on this page.
2. Preview instant results: The model detects plastic bottle, aluminum can, food wrapper, cigarette butt, paper cup 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.
3. 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.
4. 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.