Build custom computer vision apps - faster & more reliably

With Groundlight’s Python SDK, you don’t need to be a machine learning scientist to develop your own computer vision application. Groundlight’s fully managed computer vision solution takes care of the ML so you can focus on building.

Start building See the docs

AutoML

Aspects such as selecting the model architecture, choosing hyperparameters, determining the training dataset, or managing dataset labels are automated

24/7 Human Annotation

No need to label all your images yourself, Groundlight system provides annotation by humans, 24/7

Fast Edge Inference

On-premises deployment, so you can have real-time predictions without having to rely on the cloud

Groundlight integrations and compatibility

Groundlight is compatible across major development platforms and available through a REST API or Python SDK. Enjoy easy deployments using Arduino, Raspberry Pi, or any number of hardware kits.

Python SDK

With only a few lines of code, you can have custom computer vision inside your application.

Learn More

API

API to let you access your models in the cloud - no need to run your own models or hardware.

Learn More

Fast Edge Inference

We offer specialized hardware for local inference. Reduce latency, cost, network bandwidth, and energy.

Learn More

ROS

Seamlessly integrate AI-driven perception into ROS2 projects, enabling natural language queries and real-time decision-making for smarter, more adaptable robotic systems.

Learn More

Build a working computer vision application in just a few lines of code:

Code Block

PYTHON

import groundlight
from framegrab import FrameGrabber
# Initialize Groundlight client and create a Detector
gl = groundlight.Groundlight()
detector = gl.get_or_create_detector(name="doorway", query="Is the doorway open?")
# Grab an image from a camera or video stream
grabber = list(FrameGrabber.autodiscover().values())[0]
image = grabber.grab()
# Process image and get a confident answer to the Detector's query
image_query = gl.ask_confident(detector, image)
print(image_query)

Connect with us, we’re here to support you:

YouTube

Watch our tutorials and learn how computer vision can be applied to various industries.

Go to YouTube

X

Support

Reach out to us for questions and get an answer from a real human being.