fpgadeveloper/camera-fmc-vitis-platforms

Acceleration platforms for Opsero Camera FMCs

Opsero Camera FMC Vitis Platforms

Introduction

This repository contains Vitis platforms and overlays for two accelerated vision AI applications for the ZCU10x development boards and the RPi Camera FMC. These platforms and applications were ported to the ZCU10x boards from the Kria Vitis Platforms and Overlays project. Detailed information on these applications can be found in the documentation for the Kria KV260 Vision AI Starter Kit applications. These designs can be run on Certified Ubuntu 22.04 LTS for ZCU10x or PetaLinux 2022.1.

Quick Start Tutorials

The following tutorials have been written to help you to build and run these applications:

• Develop smart vision apps for ZCU106 and RPi Camera FMC
• NLP-SmartVision in PetaLinux on ZCU104 Using Raspberry Pi cameras
• Benchmarking an FPGA-based AI Vision application

Requirements

• Linux development PC with tools installed:
  • Vitis Core Dev Kit 2022.1 (Vivado, Vitis HLS)
  • PetaLinux 2022.1
• One of the supported dev boards (see below)
• RPi Camera FMC
• 1x Raspberry Pi v2 camera
• DisplayPort monitor and cable
• Ethernet cable to a network router (for network and internet access)

Supported dev boards

• AMD Xilinx ZCU106 Zynq UltraScale+ Development board
• AMD Xilinx ZCU104 Zynq UltraScale+ Development board
• AMD Xilinx ZCU102 Zynq UltraScale+ Development board

Coming soon for PetaLinux flow:

• TUL PYNQ-ZU
• Digilent Genesys-ZU
• Avnet UltraZed EV carrier

Vision AI applications

• Smartcam:
  • Runs on ZCU104 and ZCU106 boards
• NLP-Smartvision:
  • Runs on ZCU104, ZCU106 and ZCU102 boards

Hardware setup

1. Connect the target board:
   • DisplayPort monitor
   • Ethernet port to a network router
   • RPi Camera FMC to:
     • ZCU106: HPC0
     • ZCU102: HPC0
     • ZCU104: LPC
   • Raspberry Pi camera v2 to CAM1 connector of RPi Camera FMC
   • Optional: Raspberry Pi camera v2 to CAM2 connector of RPi Camera FMC

Ubuntu flow

2. Prepare SD card with Certified Ubuntu 22.04 LTS for ZCU10x
3. Boot the board, install xlnx-config snap and run xlnx-config.sysinit:

sudo snap install xlnx-config --classic --channel=2.x
sudo xlnx-config.sysinit

More complete instructions for Ubuntu flow found here.

PetaLinux flow

2. Prepare SD card with PetaLinux image (build instructions below)
3. Boot the board, build applications and run

More complete instructions for PetaLinux flow found here.

Build instructions

Clone and source tools

This repo contains submodules. To clone this repo, run:

git clone --recursive https://github.com/fpgadeveloper/camera-fmc-vitis-platforms.git

Source Vivado and PetaLinux tools:

source <path-to-petalinux>/2022.1/settings.sh
source <path-to-vivado>/2022.1/settings64.sh

Cd into the directory of the board that you want to use. For this example, we will use the ZCU106:

cd camera-fmc-vitis-platforms/zcu106

Run the make command for the flow you wish to use (Ubuntu or PetaLinux).

Ubuntu flow

The Ubuntu flow will generate a PAC which you can install and activate on your ZCU10x board.

Build all (Smartcam application):

make pac OVERLAY=smartcam

Build all (NLP-SmartVision application):

make pac OVERLAY=nlp-smartvision

The PAC will be a compressed zip file found in this directory:

camera-fmc-vitis-platforms/<target-board>/pac

PetaLinux flow

The PetaLinux flow will generate a PetaLinux project with boot files and a root filesystem that you can copy to an SD card. The root filesystem is stored and retained on the SD card.

Build all (Smartcam application):

make petalinux OVERLAY=smartcam

Build all (NLP-SmartVision application):

make petalinux OVERLAY=nlp-smartvision

The PetaLinux image files will be found in this directory:

camera-fmc-vitis-platforms/<target-board>/petalinux/<platform-name>/images/linux