What is it?

RearViewsion is a sensor fusion app that is THE missing rear view camera of a Vision Pro, based on an M5CameraX. It is a combination of hardware and software solutions for VR/AR use. The hardware/fw combination is called 写真*ShashinGo.

Why did you make it?

There is currently no way to get an image while in VR inside of an Apple Vision Pro. This seems like a glaring omission by Apple. We don't have any details around why Apple didn't enable camera image capture on the front facing cameras on a Vision Pro. We suspect there are first performance concerns, as these cameras are part of core features. But another reason is probably because of privacy. Since IoTone advocates privacy, we expect our users to honor privacy when using this camera. The hardware is based on standard ESP32 designs with basic webcams.

What makes it special?

This is a wearable AI camera. The Camera is intended to be worn on your HMD, on the strap facing backwards to allow VR users to see backwards. This is a unique capability. There are alternative configurations that can be used. The camera can be made to face the user to support streaming activity to a third party. Or this camera can face forward and be used to collect first person video streams.

Where does the "AI" come in? The answer is, it depends on which camera you choose (we have several options available), and which firmware you decide to run. The basic firmware has an AI capability for face tracking. While not incredibly useful on its own, it should open up the door for other ideas and applications.

A Vision Pro app is provided that grabs image frames into a floating window. Since the code is open, you can modify and build upon this for your own designs. IoTone will release the hardware as part of its own OpenWR (Open Wearable) initiative. +1 Open Hardware +1 Open Source. One might argue this is a missing feature of the Vision Pro.

Before Buying, Things to Consider

Well in terms of package, this is the nicest wearable form, with a cool clip on adapter, and a Lego thingy to let you hook it to lego parts. This one has USB-C. It is a variant of the M5Stack Fisheye package. The software is interesting. It doesn't run the standard "camera web server" but a simplified and clean UX. There is still a little work to get scroll and get the stream started. The firmware is ESP-IDF based, not arduino, so if you want to customize more deeply, this is a good one to start with if you are already working in the ESP-IDF ecosystem. Otherwise, if you were hoping to hack with arduino, it should be possible though we decided not to swap out this firmware since there are some interesting features in this firmware not available in the "Camera Web Server" starter. I think the camera image is cleaner.

Support

• We will offer support through our OpenWR (Open Wearable https://github.com/IoTone/OpenWR ) project.
• If you have questions about the firmware, please don't hesitate to contact us. Because there are different approaches to running your firmware, we can provide support on how to reflash your device to suit your needs.
• If you have questions regarding power, how to power, cables, connectors, or I/O, please ask. If we don't have answers, we can direct you towards the OEM as a resource for deeper hardware questions.
• Schematics and Designs: most of the hardware has schematics available. We will provide a link where possible. We don't design the original hardware for the "Concept 2" cameras, but may down the road provide custom designs in the future.

Your Purchase goes towards Open Hardware and Open Source

We mark up the prices a bit. You are free to follow the links to grab your own firmware, fork our code, and make your own builds. Totally fine. Most hardware can be found on Amazon, Alibaba, or direct from the OEM. We are focused on the integration of AI + AR/VR features. As we get more customers, we hope to take bigger steps in the improvements to the designs. Thanks for your support. Feel free to support us directly as well ( https://github.com/IoTone/OpenWR?tab=readme-ov-file#how-to-help ) using your preferred blockchain.