Starter Kits

Accelerate Oem Designs

Starter Kits

Accelerate Oem Designs

RESOURCES

At this page you find all the relevant resources to start your experience with the Deep Vision Starter Kits.

All of Deep Vision Starter Kits are based on Linux Yocto and include the Deep Vision Jumpstart Code, a source code repository with all the fundamental software elements of a vision project with accelerated AI:

  • Grabbing video with Gstreamer from UVC cameras
  • Trigger GPIO signal for camera sync
  • Gstreamer video enc/dec via NXP i.MX 8M Plus VPU
  • AI acceleration on Hailo-8™
  • AI acceleration on NXP i.MX 8M Plus NPU
  • Vision pre and post processing via OpenCV

The Deep Vision Jumpstart Code should be enough to get you started, but if your challenge requires an extra mile, we are here to support you: our talented team of engineers and computer scientists is ready to get committed on your challenge and build on top of the Starter Kits what it takes to transform images, videos or 3D data into the valuable high-level insights that you need.

As first thing, verify the kit content and check the Quickstart Guide:

  • Kit Content (v1) – LINK
  • Quickstart Guide (v1.1) – LINK

Then fill the form to request access to the Github repos:

  • Repo Access Request Form – LINK
  • Redistribution Limited License Terms – LINK

​Link to Github repositories:

  • Jumpstart code github repo – LINK
  • BSP manifest github repo – LINK
  • Yocto metalayer github repo – LINK
  • 3D+AI github repo for StereoCam Starter Kit only – LINK

Additional data:

  • Relevant resources for using the Starter Kit (v1.0) – LINK
  • Sofware components and dependencies (v1.1) – LINK