Hacker News Reader

Comment by Stanleyc23 | Hacker News Reader

There was also a very brief mention of a "depth api" before they started talking about the actual ARKit. is that endpoint documented anywhere?

View on HN

I think you're referring to this, depth data mapping from the dual camera in the iPhone 7+:

https://developer.apple.com/documentation/avfoundation/avdep...

here? https://developer.apple.com/documentation/arkit

Nah, it was during the camera bits, they noted that the dual camera lets them add depth querying/manipulation and they're providing an API for it.

nah I think its something for normal photos, not immersive content.

> Camera: Updated video and image compression, improved Portrait Mode with support for low light, plus a new Depth API for developers.

from https://venturebeat.com/2017/06/05/apple-announces-ios-11/

I had a look at the documentation at [1], which says "The AVCaptureDepthDataOutput class captures and delivers depth data in a stream (similar to how the AVCaptureVideoDataOutput delivers video data)."

So you can now access a stream of depth data. I'd put money on ARToolkit using this stream when it's available. For now, it would only be from the iPhone 7+.

[1] https://developer.apple.com/documentation/avfoundation/avdep...

AR kit isn't limited to the 7+ ( they demoed it on an iPad) space so clearly that's not REQUIRED. I wonder if they're using the second camera for depth on the 7+ when it's available.

Perhaps that data isn't granular enough (or fast enough) to be useful when doing AR compared to the normal camera picture plus the motion sensor data.

ah great find.

Reply on news.ycombinator.com