Welcome to a new video, in this episode I’m gonna show you how to install frigate stand alone on a single board computer running Ubuntu, how to configure it to start recording on an external usb drive with person detection enabled.
For those of you who haven’t heard about Frigate, it’s an open-source NVR with real-time AI object detection built-in, all recording and processing happens locally on your own hardware so a good thumbs up for being open-source, we love open-source projects on the voltlog channel but another thumbs up is for hosting everything local, so you have security & privacy under your own control.
You can also run Frigate on your Homeassistant server if you wish to do so, but some people might opt to run it independently on another computer for increased processing power and more storage options.
For this tutorial I’m gonna be installing it on the Khadas VIM4 single board computer and this thing is a beast. I’m calling this the Pi Killer for reasons which I will outline in a few moments. This was released this summer, it’s the latest model in Khadas lineup and comes with some stellar specs: This has an 8 core Amlogic A311D2 processor, four of the cores are ARM Cortex-A73, running at 2.2GHz, the other four are Cortex-A53 running at 2GHz and there is an extra 32 bit STM32 microprocessor on here, There is a powerful Mali G52MP8 GPU which supports up to 8K decoding, we get 8GB LPDDR4X RAM on this board, 32GB of eMMC storage, 32MB of SPI flash memory which runs it’s own custom bootloader called OWOW, Bluetooth & Wifi 6 connectivity plus 10 gigabit ethernet, there is an M.2 slot for SSD on the back and the list goes on and on, they practically jammed everything that’s nice on this board and out of the box it can run Ubuntu & Android.
Now you’re probably asking yourself how does this stackup against the Raspberry Pi4 so here is a simple comparison just based on specs, if we look at processor power, ram memory, storage options, connectivity the VIM4 clearly has more options and more horsepower than a Pi4, the only place where the PI4 has more options is with USB connectivity. But having all of this extra horse power does come at a cost and the VIM4 can get pricey especially if you order it with an active cooling kit which is not included by default but you will need if you plan to benefit from that processing power. For the latest price information please check out the links I’ve placed in the description of the video. We could also talk about availability with the VIM4 being available for order right now, while the Pi4 being harder to find. Due to the VIM4 processor being manufactured on a 12nm process we can also expect better power efficiency, providing more performance per watt over the Pi4.
Another impressive feature of the Amlogic processor is the built-in Neural Processing Unit which is rated for 5 TOPS. By comparison a google coral is only rated for 4 TOPS but as we all know a google coral is well supported over many different interfaces while the new NPU inside the Amlogic processor is there but not actually supported in the software due to license requirements which are not yet sorted.
List of CLI commands used for this setup
#Update the apt package index and install packages to allow apt to use a repository over HTTPS:
sudo apt-get update
sudo apt-get install \
ca-certificates \
curl \
gnupg \
lsb-release#Add Docker’s official GPG key:
sudo mkdir -p /etc/apt/keyrings
curl -fsSL https://download.docker.com/linux/ubuntu/gpg | sudo gpg –dearmor -o /etc/apt/keyrings/docker.gpg
echo \
“deb [arch=$(dpkg –print-architecture) signed-by=/etc/apt/keyrings/docker.gpg] https://download.docker.com/linux/ubuntu \
$(lsb_release -cs) stable” | sudo tee /etc/apt/sources.list.d/docker.list > /dev/null#Install Docker Engine
sudo apt-get update
sudo apt-get install docker-ce docker-ce-cli containerd.io docker-compose-plugin docker-compose#start docker
sudo systemctl enable docker
sudo systemctl start docker#add docker group
sudo groupadd docker
sudo usermod -aG docker $USER#reboot to apply changes
sudo reboot now