Skip to main content

Project Aria Client SDK and CLI Setup Guide

Overview​

The page provides instructions about how to get started with the Project Aria Client SDK, covering:

Requirements​

Hardware​

  • Project Aria glasses that have:
  • If you want to stream over Wi-Fi, you'll need a router, such as Asus, Netgear or TP-Link, that has:
    • No firewall
    • Supports Wi-Fi 6
      • So that the glasses can connect to the 5GHz band when streaming over Wi-Fi
danger

The Client SDK does not currently support streaming over corporate, university or public networks. Those networks are protected by many layers of security and firewalls. We recommend using one of the recommended routers listed above to stream over Wi-Fi.

Platforms​

The codebase is supported on the following platforms:

  • x64 Linux distributions of:
    • Fedora 36 or newer
    • Ubuntu jammy (22.04) or newer
  • Mac Intel or Mac ARM-based (M1) with MacOS 11 (Big Sur) or newer

Software​

  • Python 3 with versions 3.8.10+
    • Python 3.9+ if you want to use the device_stream code sample due to the fastplotlib dependency or if you want to use Jupyter notebooks
    • Python 3 download page
    • To check which version of Python 3 you have, use python3 --version
  • ADB (optional)
    • In addition to the CLI, you can use ADB to interact with Aria glasses
    • ADB is one of the ways that you can download Aria data

Environment Setup​

Step 1: Install SDK from PyPI​

Create a virtual environment​

When using pip, it is best practice to use a virtual environment. This will keep all the modules under one folder and will not break your local environment. Use the following command with your version of Python3.

python3 -m venv ~/venv

Install the Client SDK and CLI​

Install projectaria_client_sdk with pip

source ~/venv/bin/activate

python3 -m pip install projectaria_client_sdk --no-cache-dir

Step 2: Run Project Aria Doctor utility​

The Project Aria Doctor utility can help detect and resolve common issues connecting and streaming from the glasses.

Run the utility and follow the prompts to resolve any issues.

aria-doctor
info

If you're on MacOS and lose internet connection while streaming, run aria-doctor again.

Step 3: Pair Aria Glasses with your computer​

Pairing your Aria glasses to your computer allows the Client SDK and CLI to control the glasses. A pair of Aria glasses can be paired to multiple computers.

  1. Turn on your Aria glasses and connect it to your computer using the provided USB cable
  2. Open the Mobile Companion app on your phone
  3. On your computer, run:
aria auth pair
  1. A prompt should then appear in the Mobile app, tap Approve to pair your glasses
    • The hash in the terminal and the app should be the same
    • View (or revoke) certificates by going to Device Settings
    • The Client SDK Certificate will remain valid until you manually revoke it or factory reset your glasses
Companion App Client SDK pairing
info

At this point, you can now use the Aria CLI to interact with you Aria glasses.

Step 4: Extract and explore the sample apps​

  1. Extract the Client SDK code samples (here to your home directory)
python -m aria.extract_sdk_samples --output ~
  1. Navigate to the sample folder
cd ~/projectaria_client_sdk_samples
  1. Install necessary dependencies:
python3 -m pip install -r requirements.txt

All code samples can be run from cd ~/projectaria_client_sdk_samples, with the exception of the Time Synchronization TICSync code samples. To use TICSync, go to cd ~/projectaria_client_sdk_samples/ticsync.

Go to Code Samples to explore Aria Client SDK features.

Go to Streaming Internals to understand how streaming works and how to configure your own streaming setup.

Go to Time Synchronization for how to create recordings with multiple Project Aria glasses.

If you encounter any issues please run aria-doctor in a separate terminal or check out troubleshooting.

info

You can check your Aria glasses' recording or streaming status in the Mobile Companion app.

danger

The Client SDK does not currently support streaming over corporate, university or public networks. Those networks are protected by many layers of security and firewalls. We recommend using one of the recommended routers listed above to stream over Wi-Fi.