Bringing Deep Learning Computer Vision to the Edge

We are providing developers a simple and easy-to-use platform to build and deploy computer vision applications on embedded devices.

 

APPLY FOR OUR PRIVATE BETA
Computer Vision Use Case

1

Choose a Computer Vision Use Case

Detect, classify and track objects, including people.

Download Middleware

2

Select a Pre-trained Model

Choose from our catalogue -- or provide your own model.

Use the SDK to Build a CV App

3

Use the SDK to Build a CV App

Create a customized AI solution to meet your business needs.

Deploy Computer Vision Applications

4

Deploy to Embedded Devices

Deploy your application on the ARM board of your choice.

Our Product Stack – Complete and Integrated

The alwaysAI Deep Learning CV Platform provides you with the following: 

  • Core computer vision services such as object detection and classification, semantic segmentation, and object tracking exposed as consistent Python APIs
  • A catalogue that standardizes Caffe and TensorFlow models across 12 key parameters
  • Middleware that integrates over 50 language libraries and deep learning frameworks compiled from source code and optimized for the ARM Cortex
  • Linux kernel and device drivers optimized for specific system on a chip (SoC) platforms as well as container images to support a reliable development environment across a variety of resource-constrained and unconnected edge devices
  • End-to-end versioning and dependency management

 

 

alwaysAI Computer Vision Platform Product Stack

Apply for the Beta Now

As a participant, you get early access to our platform and an open channel of communication with our team.

The alwaysAI team is continuously adding functionality to the platform for its broad market release later this summer. And we are super excited for the developer community to build powerful computer vision applications on the edge. We hope you’ll join our beta program!

What you’ll need:

  1. An ARM32- or ARM64-based developer board running Linux, such as Qualcomm Snapdragon, NVIDIA Jetson Nano, Google Coral, or Raspberry Pi 3 b+.
  2. A 1080p USB webcam
  3. A few hours in July to create your first project with our platform