Tutorial post: Doing Photogrammetry solo

In this post, I’ll be going over the ins-and-outs of doing photogrammetry on your own, and I how I messed up my group’s first model attempt using an archive object.

I. The setup:

First and foremost, you need your own set of photos to create the model via pixel models. In this class, we’re all using the educational license—so, similarly to how we navigated building our dataset for the museum archive project, we want between eighty and one-hundred photos of various angles at a close distance.

A lesson I learned, very quickly, was that my setup for this project was very poor. I took for granted the previous setup provided to us by Special Collections; a dark room that included an easily-strafed pedestal and balanced lighting to minimize glare. The setup I used was a large table covered with photography paper, which I later learned was not sufficient; I entrusted the algorithm more than I should’ve with being able to distinguish between the object and the background. I’ll elaborate further later on.

II. .HEIC to .jpg conversion:

The photos I took were all the default High Efficiency Image Capture format from IOS, which needed to be ported into .jpg format to be usable with the AutoDesk Photo Recap software. The app I recommend for this is iMazing Converter, a free tool that allows for easy automatic file conversion.

III. AutoDesk suite apps:

The three AutoDesk apps needed for this process are Autodesk ReCap, ReCap Photo and Autodesk Desktop Connector. Installing the three is time-consuming and requires attention, as ReCap Photo is only available as a download extension (an option you need to actively opt into) in order to install. In order to use the cloud computing included in the Autodesk suite, you need to have AutoDesk Desktop Connector active, which will run in the background—allowing your project to remain in the processing queue even if you’re not at your computer.

IV. Creating an object model:

Assuming you have Recap Photo installed on your computer, and you’ve verified your account is logged in and is using an educational subscription, you’re ready to make an object project. Here’s where your converted files come in—click on the option pictured above and in the interface that follows, highlight your eighty to one-hundred photos you converted earlier and drag them into the window like so:

Before hitting create, I recommend saving the output to a location that’s easy to remember and simple to navigate to. Once you’ve done so, hit the blue ‘create’ button and your model will enter the cloud processing queue. As I mentioned earlier, you don’t need your computer to be open once the uploading process is complete—having desktop connector open ensures that you stay connected. There is a downside to this process: it can take anywhere between a half hour to five, so make sure you’ve got other work to do to keep yourself occupied while you wait!

V. Opening and viewing your project:

Once the project is done processing, Recap Photo will allow you to open it under the “projects” section:

Opening the project is as simple as clicking the displayed .rcm file, which lets you see:

An absolute mess because my setup for the photos was terrible! The most important part of photogrammetry is taking good photos, which I did not. This run was, entirely, a proof of concept that I could at least get the steps to work. If you’d like to know more about setting up your own photogrammetry setup, here’s a great article by Paleontologist Heinrich Mallison that provides some tips and tricks he uses to create 3D models of dinosaur fossils (almost all of them being irregularly shaped) in his own office. Photogrammetry is fairly user-friendly, but how well your final product turns out, I’ve learned, is entirely dependent on user competency!