Roboflow Features

Label images using Roboflow Annotate. Upload and export JSON, XML, CSV, and TXT annotation formats.

Upload JPG, PNG, and BMP image files.

Use your Roboflow Train models to accelerate your annotation work by using AI as a starting point to apply labels.

Upload video files and Roboflow will extract frames based on the sample rate you specify. Upload mov, mp4, avi files.

Preprocessing: Auto-orient, resize, grayscale, auto-contrast, static crop, tile, class modification, & filter null.

Organize your data, your way. Create as many datasets as you need without worrying about limits.

Choose from 15+ different formats when exporting your training data and refer back to previous versions to track your progress.

For each export, get a curl link that can be pasted directly into your notebooks and download from the command line.

Keep datasets securely stored and archived indefinitely.

Divide your data into sets (train, validate, and test) so you have everything you need to evaluate your model performance after training.

Health Check: Detailed statistics and charts containing helpful insights into the quality of your dataset and how to improve it.

Train your model with one click using Roboflow Train. Within 24 hours, you'll get results including mAP, precision, and recall, as well as a hosted API for inference.

Direct integrations with hosted computer vision model training services including Amazon Rekognition, Google Cloud Vision AutoML, Azure Custom Vision.

Hosted API to receive predictions from your model as fast as your network connection can handle sending the frames.

Autoscaling infrastructure includes load balancing, supports burst and no burst, and is always on without any custom engineering work.

Collect sample inferences whether at specific time intervals, at random, and with different confidence thresholds to improve your model's performance

roboflow.js web, NVIDIA Jetson, Luxonis OAK, Mobile (iOS), self-hosting, VPC and on-premise.

Run your model on embedded devices for drones, robotics, IoT, offline scenarios, and more.

Use Luxonis OAK devices via pip install roboflowoak or via Docker for a standardized device that combines a camera with a built in hardware accelerator

Mobile iOS SDK is available for deploying your model directly to an iOS application running on iPhone or iPad.

Optimized it to get maximum performance from the NVIDIA Jetson line of edge-AI devices by specifically tailoring the drivers, libraries, and binaries specifically to its CPU and GPU architectures.

Deploy our Docker container in your private cloud (or on your own metal) using our Inference Server and our License Server Docker containers. Cache weights for up to 30 days to run completely air-gapped or in locations where an Internet connection is not readily available.

Roboflow holds SOC 2 Type 2 certification for our security and data privacy standards.

CVAT, VoTT, LabelImg, LabelMe, RectLabel, VGG Image Annotator, Scale, Labelbox, SuperAnnotate, Google Human Labeling, Amazon SageMaker Ground Truth, and more.

Autoscaling infrastructure includes load balancing, supports burst and no burst, and is always on without any custom engineering work.

Share datasets with other members of your team and seamlessly collaborate on projects.

A comprehensive set of resources to get you started in computer vision with the help of Roboflow’s tools for preparing and managing your datasets.

Developer documentation walking you through how to use Roboflow and how to get the most out of our tools.

Can't find the answers you need in our documentation or on the Roboflow Blog? Join the Roboflow forums for help from the Roboflow team and community.

Helpful tutorials on creating and training models along with interviews with leading computer vision experts.

Your questions and requests will be bumped to the top of the queue and answered by computer vision experts on our team via email and in-app live chat.