Build with RF-DETR
Build a Package Counting App Free with RF-DETR
Count every box and package moving down a conveyor, each one exactly once. Label footage, train RF-DETR free, and deploy it as a workflow.
- No credit card
- No ML team required
- Edge, on-prem or API
Build a Package Counting App with Roboflow’s Free RF-DETR
In short
A package counting app uses computer vision to detect boxes and parcels in camera video and count each one once as it passes a set point on a conveyor or dock. The model to train for it is RF-DETR, Roboflow's open-source real-time detection model, paired with a tracker and a line counter. With Roboflow you label your footage, train RF-DETR for free, and deploy it as a workflow that returns a running count, annotated frames, and tracked detections.
How it works
The workflow behind the package counting app
Use this Package Counting with RF-DETR workflow: each frame goes to an RF-DETR model that finds boxes and packages, then through a confidence filter to a ByteTrack tracker that gives every package an ID. A line counter counts each tracked package once as it crosses, and the boxes, IDs and running count are drawn onto the frame. Every block is editable in the low-code builder, and the Roboflow Agent can build it from a description.
- output_image
- package_count
- tracked_packages
How to build a package counting app with RF-DETR
Open the workflow and add your footage
Click "Use this free template" to copy the workflow into your Roboflow workspace. Upload frames from the camera over your conveyor, with the lighting, package mix, and belt speed you run.
Label packages and train RF-DETR
Use Auto Label to draw boxes for box and package, then review and correct them. Train an RF-DETR model on the labeled frames and select it in the detect_packages block. RF-DETR fine-tunes well on small custom datasets, so you can start with the footage you have.
Set the counting line and thresholds
Move the line in count_packages to the point where packages pass your camera, so each one is counted as it crosses. Adjust the thresholds in filter_by_confidence for your package mix.
Deploy and connect it to your operation
Deploy the workflow as a Serverless Cloud API, or on your own hardware through Inference, to run on live camera streams. Each frame returns output_image, package_count, and tracked_packages for your warehouse system or dashboard to use.
Roboflow MCP
Build it from your AI assistant with the Roboflow MCP
Connect the Roboflow MCP Server to Claude, Cursor, or any MCP-compatible assistant and build the whole package counting app in natural language, without leaving your editor.
Ask an assistant such as Claude or ChatGPT to search Roboflow Universe for package datasets, upload and auto-label your conveyor footage, and train an RF-DETR model on box and package classes. It can then create and run the package counting workflow on test frames and deploy it to your devices.
Paste it into your coding agent and it walks you through the rest. Setting up by hand? The server is https://mcp.roboflow.com/mcp.
Open the Roboflow listing, select Connect, then sign in to Roboflow. Available on desktop, web and mobile.
Run these commands in your terminal, then sign in to Roboflow. Claude Code CLI has no documented browser install link.
Opens the Roboflow plugin listing. Sign in if prompted and confirm the install on Cursor to add MCP and skills.
Opens the Roboflow plugin listing. Install the plugin and sign in to Roboflow when prompted to connect your account.
Opens the Roboflow plugin listing shared by ChatGPT and Codex. Install the plugin and sign in to Roboflow when prompted to connect your account.
Then run /mcp auth roboflow in Gemini CLI to sign in. Gemini CLI has no documented browser install link.
What your package counting app can do
Trained on your footage, an RF-DETR model handles the specific, hard-to-catch cases that matter on your dock:
- Count packages that touch or overlap on the belt as separate items
- Count each package once, even if it pauses or backs up
- Handle boxes, poly mailers, and envelopes of every size
- Keep counting through changing light at dock doors
- Ignore labels, tape, and shadows that look like edges
- Keep pace with fast-moving sortation lines
Want to customize your package counting app?
Describe the package counting app you want in plain English: the camera view, what you are counting, and where to count it. The Roboflow Agent finds the data, trains an RF-DETR model, and assembles the workflow for you.
Open source
Roboflow’s RF-DETR is open source, Apache-2.0, and free to use
RF-DETR is an open-source, Apache-2.0 licensed real-time object detection model developed by Roboflow. The rfdetr Python package and most RF-DETR models are free to use, including in commercial products, and you can train and run everything on your own hardware. No account or lock-in required.
Apache-2.0 and free
The rfdetr package, Nano to Large detection, every segmentation size and Keypoint are Apache-2.0 licensed and free to use, including commercially.
Install and train locally
pip install rfdetr on Python 3.10+. Fine-tune on your own data in a few lines and run it entirely on your machine.
State of the art on COCO
RF-DETR is a real-time detector that is SOTA on COCO. RF-DETR-Large reaches 56.5 AP at 6.8 ms on an NVIDIA T4, and uses a DINOv2 backbone.
Deploy anywhere
Export to ONNX or TensorRT and run on GPU, CPU, or NVIDIA Jetson at the edge, or serve it through the Roboflow hosted API.
| Models | Licence |
|---|---|
| Detection: Nano, Small, Medium, Large | Apache-2.0 |
| Segmentation: every size, Nano to 2XLarge | Apache-2.0 |
| Keypoint (preview) | Apache-2.0 |
| Detection: Atto, Femto, Pico, XL, 2XL (rfdetr[plus]) | PML 1.0: commercial use included on Roboflow; commercial self-hosting needs the Enterprise add-on |
RF-DETR vs. other free detectors
A package counting app has to detect every package on every frame, fast enough to keep up with a moving belt. Here is how RF-DETR, fine-tuned on your own conveyor footage, compares for real-time counting.
| Model | Tasks | Best for | License |
|---|---|---|---|
| RF-DETR | Detection, segmentation (keypoints in preview) | Real-time detection you fine-tune on your own data, one model across tasks | Apache-2.0 (Nano to Large, all segmentation); PML 1.0 (Atto, Femto, Pico, XL, 2XL) |
| D-FINE | Detection | Precision localization when you do not need to fine-tune | Apache-2.0 |
| YOLO | Detection, segmentation, pose | Fast CNN-based detection with a large ecosystem | AGPL-3.0 (commercial license required for closed-source use) |
| GroundingDINO | Open-vocabulary detection | Finding objects by text prompt with no training, not real-time | Apache-2.0 |
| SAM 2 / SAM 3 | Segmentation | Pixel-precise masks; SAM 3 adds text-prompted segmentation | Apache-2.0 (SAM 2), SAM License (SAM 3) |
| Florence-2 | Vision-language | Reading, captioning, and OCR on images | MIT |
The takeaway for this app: RF-DETR runs in real time, so it keeps pace with a moving conveyor and gives the tracker clean detections on every frame. It fine-tunes well on small custom datasets and is open source, so you can train on your own packages and run the model wherever your cameras are.
Build your package counting app free today
Label, train, and deploy your package counting app in one place.