If you haven’t already seen how expertly I was able to 3D print my head to create the perfect headphone stand for his new pair of headphones, then do yourself a favour and check out the roughly three-minute short below.
Once you have seen the video, or if you’re one of the cool kids who watched it previously, this article will get into the nitty-gritty of how we got it right and a few extra things I just couldn’t fit in the short three minutes that social media provides us these days.
Why we chose the Kiri Engine
First, let me explain why we used Kiri Engine over the more popular Polycam.
Polycam limits us to twelve scans before we have to pay for it, whereas Kiri Engine allows us to scan unlimited objects. The limitation is that we can only export three of these scans weekly. This gave us the freedom to get a little bit of practice before actually shooting the video. Looking at the monthly subscriptions for both Polycam and Kiri Engine, Polycam will set you back $25 (around R470), whereas Kiri Engine goes for $18 (around R340).
When paying, the limitations of scans and exports are removed, and you gain access to all the features. While both apps offer similar uses, there are two main differences between them. Polycam has been around the block a lot longer than Kiri Engine. This means that their app is slightly better in terms of usage. Things like saving the raw data and scan processing are much easier to do in Polycam than Kiri Engine, with Kiri Engine adding a couple more steps to achieve the same results. Polycam also allows you to use all of the cameras on your phone, whereas Kiri Engine only supports main camera usage. This can be circumvented by taking the photos manually and then uploading them to Kiri Engine, but again, more steps are needed to achieve the same results.
In terms of data processing, Polycam allows up to 2,000 photos to process your scan, whereas Kiri Engine is limited to only 200. In terms of data output, Polycam exports in 8K resolution, whereas Kiri Engine only goes up to 4K. This will affect users who want to keep textures on their scans for use in video or other types of editing. Kiri Engine performs better than Polycam when it comes to producing polygons. Roughly speaking, the Kiri Engine produces almost twice the amount of polygons that Polycam does. This results in a more detailed 3D scan when the textures are removed. For 3D printing purposes, this is usually better.
As we didn’t want to commit to a subscription just yet, I loved that Kiri Engine offered me a better, free-to-use option for what we wanted to achieve.
For my first attempt, I tried using the manual photo option to complete a scan. This method allows the user to take a substantially large amount of photos of every angle that they think needs to be covered. My first attempt was ok, but it’s a lot of work to do it manually and the cube I scanned came out as more of a blob anyway.
Luckily, because I had an iPhone in my pocket, I was able to try out LiDAR scanning to see if it made a difference. It did.
As iPhones are currently the only phones that pack a LiDAR scanner, this feature (both in Polycam and Kiri Engine) is only available to iPhone users. Android users will have to stick with Photogrammetry for now. This also means Android users will probably benefit more from Polycams larger number of accepted photos than Kiri Engine’s 200.
To explain the difference between LiDAR and Photogrammetry, LiDAR uses laser pulses to detect the object’s depth and distance by measuring the pulse’s time of flight. Photogrammetry uses layers of photographs to build the image from all angles. Photogrammetry, in most cases, should provide you with a more accurate scan as it can produce life-like and accurate models based on photos, whereas LiDAR produces a point cloud that has less data about the surface of an object and more about its shape, distance and size.
Once you’ve opened LiDAR mode, the app will ask you to position the object inside a virtual, on-screen cube. This allows the app to understand the position of the object in question and then prompts you to rotate around the object, ensuring you have all angles covered. As you slowly move around the object, which was my head in this case, the app dings every time it’s happy with that angle and asks you to move on. This was much easier than manually capturing my head into the app. Kiri Engine does throw an on-screen pop-up if it thinks there is not enough light in the scene. This helps in some instances, but as we saw with the sunlight, too much light can also be a problem. So try not overdo it when using the LiDAR mode.
In our first attempt, I stood outside in the sunlight for scanning. The 3D object came out better than the blobby cube, but once the processing was completed, there was still a divot in the back of my head. I assume the divot was created due to too much sunlight being on that specific spot. Perhaps the light affected the LiDAR scanning, and the app wasn’t properly able to detect the depth of the back of my head.
Our second run was done inside the studio, with our studio lights set up and positioned to provide a more even light across my head and the room instead of direct light pointing straight at me. This time, the result worked out better and was the scan we used in the video.

Exporting was as easy as hitting the export button and sending the file via WhatsApp. The file was downloaded onto a PC and opened in Creality Print.
Editing
We were pretty impressed by the quality of the scan, so we opted not to open the file in Blender for extra smoothing or gap covering – I’m sure I would’ve ruined my face even more in Blender anyway.
The scan included a bit more of my shoulders than we wanted, so we sliced the object just below the throat and deleted the bit we didn’t want.
Then, using Len’s patented hand measurement technique, we measured the size of my head. We adjusted the 3D object in Creality Print to these measurements to get the print as close to life as possible and decided that would be that. You’ll later read that this measuring technique didn’t work as expected. In future I would recommend we use a tailor’s tape. This will allow us to measure around my head more accurately for better scaling options.
After spinning the object around in the application for a bit, I thought my huge nose might topple the completed print over, so we added a base for stability. We threw in a base that some of you may recognise from last year, chopped off the bottom of it and stuck it just underneath the head. After some upscaling and a bit more spinning around, we were happy.

Printing
To save on filament, I adjusted the head’s infill to 1%. I wouldn’t normally go this low, but an object of this size can almost use up a whole roll of PLA. The print came out a little fragile, but with careful handling, my head should survive. I suggest adjusting the infill to at least 10% for good, real-world stability.
Auto tree supports held everything in place, and the object was sent to print. The handy ethernet camera on the Creality K1 Max provided hourly checks on the print, and helped ensure the print came out right the first time.
We had no issues using Creality’s PLA+, in red, for this print. The print consumed just over 500g of filament, saving more than 300g because of the infill adjustments.

Post Print Processing
Once the print was complete, it was time to remove the support and sand the model down. The supports from the head came off pretty easily, while the tiny trees at the bottom of the base proved more difficult. Next time, these tiny supports can be avoided by adding a proper, flat base instead of the one we used for this print. Either way, they came off okay in the end.
Once removed, the sharp corners and leftover bits from the supports were sanded down using my nifty electric nail file. The electric nail file I use is just a cheap device I picked up from my local pharmacy. It provides a few different heads that can be added for sanding down sharp corners and smoothing over small errors in the print. It’s essentially a Dremel tool for those of us on a budget. We picked up a cheap set from Clicks for about R180. This thing has saved my butt on more than one occasion, and I implore anyone doing 3D printing with an FDM printer to get one. Or a Dremel, if you want to be fancy.

I also decided that the headphone stand would look a lot better with a lick of paint. So I base coated it in a nice matt black colour using Rust-Oleum Painter’s Touch Plus Primer Spray. Avoid the one that includes the paint as well; this messes things up for any future layers of paint you want to throw on your print. I then speckled the print with some white spray paint to give it a galaxy-looking effect and painted the NAG logo in lovely gold.

The paint made a big difference, and if you have the time, be sure to give your prints a little love with a splash of colour.
And that was pretty much it. It’s much easier than it sounds, so don’t be scared to try it alone. If you run into a bit of trouble, though, feel free to reach out to us, and we’ll do our best to guide you down the path of printing your own head.
We’d like to thank our friends over at 3D Store for supplying the Creality K1 Max that we used to print this glorious head, and you should check them out if you’re in the market for any 3D printing hardware or accessories.