Hardware · build

A $5 Vision Pro clone built on a phone you own

brenpolyon GitHub

CardboardXR turns an old-school phone headset into 6DOF tracking, camera passthrough, hand tracking and gesture control. The catch: you need a flagship phone already in your pocket.

Build sheet

Device: Flagship phone (Pixel 9 Pro, iPhone 15 Pro)
Headset: 3D-printed Cardboard-style shell
Software: Unity + Google Cardboard SDK
Tracking: 6DOF, passthrough, hand tracking

Can I build this?Beginner · AI 4/5 · $5–25

Difficulty: Beginner
AI-buildability: 4/5
Time: A weekend
Cost: $5–25

Parts

Flagship phone (Pixel 9 Pro or iPhone 15 Pro)3D-printed headset shell VR lenses

Tools

3D printerUnityAndroid Studio or Xcode

Starter prompt

Based on "A $5 Vision Pro clone built on a phone you own" on Afterware: https://afterware.xyz/p/a-5-vision-pro-clone-built-on-a-phone-you-own

Help me build a phone-based VR headset in Unity that uses the phone's 6DOF tracking for head pose, shows camera passthrough in stereo, and runs hand tracking through the built-in ML models. The headset is a 3D-printed Cardboard-style shell with adjustable lenses. Start with a Unity scene that renders split-screen stereo and reads the phone's pose each frame.

Estimates based on the original write-up. How AI-buildability is scored. Building with an AI assistant? Connect it to Afterware (free MCP server).

Story

Google killed mobile VR, but [brenpoly]'s CardboardXR argues phones have gotten good enough to do the job anyway. His project video calls it "A 5$ Apple Vision Pro." The $5 figure assumes you already own a flagship like the Pixel 9 Pro or iPhone 15 Pro he tested with. Those phones give him 6 degrees of freedom tracking, camera passthrough, object detection, hand tracking and gesture control, the features that used to be 3DOF and zero passthrough on mobile VR.

The headset is a 3D-printed replacement for the old face-hugging cardboard, with adjustable optics. The printed shell is sold on Makerworld. The lenses are still available online, and those are what he specified. The software is built in Unity, which brings the VR/AR toolkit and the machine learning models for hand tracking, and it draws on the Google Cardboard SDK that Google open-sourced when it stopped development.

If you have a flagship phone sitting unused, this is a cheap way to find out what current phone sensors can do strapped to your face. The GitHub repo and video are the place to start.

Resources · 2

CardboardXR GitHub repo Project video