The first app able to use Apple’s new Object Capture API has appeared: PhotoCatch.
PhotoCatch runs on Mac OS and leverages Apple’s recently announced Object Capture API, which is part of the upcoming Mac OS Monterey release. While this OS upgrade is likely to be released to the public in October, it’s possible to load the beta test version on a suitably equipped Mac to test PhotoCatch.
Object Capture is a service built into Mac OS Monterey that performs highly optimized photogrammetry on a collection of supplied images. The photogrammetry algorithms use differences in the images to determine the location in 3D space of each point, and from this a 3D model can be derived.
The process of using Object Capture is a bit different than with a traditional 3D scanner. The procedure is to use an iPhone equipped with a depth camera in Portrait Mode to take a series of images of a subject from all angles. Using Portrait Mode allows the API to isolate the subject against the background and surrounding elements.
These captured images must then be submitted to Object Capture on Mac OS, the second step in the process. Heavy processing occurs to generate the 3D models, which Apple primarily intends for use in AR / VR applications. However, the 3D models can also be used for 3D printing and incorporation into more complex 3D models.
Apple provided a sample Object Capture source code application in their Mac OS Monterey system development kit, but it had to be compiled and installed manually on Mac OS by a developer. We tried that in our previous coverage, but that’s something that few Mac users can do on their own.
PhotoCatch For 3D Printing
Enter PhotoCatch, which is the first purpose-built app to use Object Capture.
It’s a relatively straightforward app, as it simply accepts a collection of images and triggers the Object Capture API. It’s easy enough for anyone to use, and that’s the app I’ve been looking for.
There’s another catch, however. Along with the requirement to be running the beta version of Mac OS Monterey, the Mac on which it runs must be relatively powerful. PhotoCatch requires either a spiffy new M1 Mac, or an Intel-based Mac with at least 16MB and a 4GB AMD graphics card. That limits use of PhotoCatch to only the latest set of machines or the most powerful from recent years.
I suspect at this point you’re discouraged. If you don’t have the right Mac hardware you’re out of luck. Even if you do, you might not want to install beta software on a machine on which you do your daily work.
But rejoice, there is an answer: PhotoCatch has a cloud-based solution, PhotoCatch Cloud.
PhotoCatch Cloud allows you to upload a folder containing images (and they don’t even have to be depth-enabled) for processing. They say you can “receive a 3D model in under 24 hours”, after “payment received”.
The payment seems to be US$1.99 per scan, which is actually pretty decent for casual users.
I think PhotoCatch is only the beginning, as we’ll see in coming months even more sophisticated apps that use Object Capture. I can imagine using a mobile app that guides you through the imaging sequence, and automatically sends the images to your Mac for processing. That would make the entire process almost seamless.