Requirements

In order to test this design on hardware, you will need the following:

  • Vivado 2025.2

  • Vitis 2025.2

  • PetaLinux Tools 2025.2 (for the PetaLinux flow), or Google’s repo tool (for the Yocto flow)

  • Linux PC or Virtual machine (for the Linux builds)

  • 1x Hailo-8 M.2 AI Acceleration Module

  • One or more Raspberry Pi Camera Module 2

  • 1x RPi Camera FMC

  • 1x FPGA Drive FMC Gen4 (for zcu106 target design only)

  • 1x M.2 M-key Stack FMC (for all other designs)

  • One DisplayPort monitor that supports 1920x1080 at 60 Hz (the design outputs this mode)

  • Alternatively: an HDMI monitor that supports 1920x1080 at 60 Hz and a DP-to-HDMI adapter

  • A microSD card (8 GB or more) and a card reader

  • Optional, on the zcu106_hpc0 and uzev targets only: a second Hailo-8 M.2 module or an M.2 NVMe SSD for the second M.2 slot

  • One of the supported carrier boards listed below

List of supported boards

Zynq UltraScale+ boards

Carrier board

Supported FMC connector(s)

ZCU104

LPC

ZCU106

HPC0+HPC1 HPC0

PYNQ-ZU

LPC

UltraZed-EV Carrier

HPC

For list of the target designs showing the number of cameras and M.2 slots supported, refer to the build instructions.

Supported cameras

The RPi Camera FMC is designed to support all cameras with the standard 15-pin Raspberry Pi camera interface, however these example designs currently only have the software support for the Raspberry Pi Camera Module 2.

Tip

We’re working on developing software support for more cameras. If you’d like to help with this effort, your pull requests are more than welcome.