Note: This article was updated on 20/10/2018 to reflect changes in software used.
On my quest to introduce 3D applications into the world of Egyptology, I stumbled upon this video, which shows how to 3D scan an object using just a camera and free software (Visual SFM, CMVS and MeshLab). I tried this method, but unfortunately, my former laptop got stuck using MeshLab. In another video, an even simpler method is explained using Autodesk Memento (later called ReMake, currently ReCap). Autodesk is the producer of famous 3D modelling software such as 3DS Max and Maya. (Note that they provide free educational licenses!)
Using Memento, it was easy as pie to create a 3D model using a camera and free beta version of the software. You simply take overlapping photos from multiple angles all around the object and feed them into the program. The Memento server does the rest, and you’re all set to upload your model to the Memento gallery (no longer in use) or for example to the online 3D platform Sketchfab. As the video explains, it is best to photograph in diffused daylight (makes for less shading), turn off your camera’s autofocus and manually focus once, taking two pictures from every angle. In my enthusiasm I simply photographed inside, in a quite unsystematic way, and taking only about 35 pictures, but even then the result is nice. This is a souvenir head I bought at Deir el-Medina:
This technique can be used not only to 3D scan small objects, but also larger objects and even complete rooms or environments (temples, tombs), as long as a sufficient amount of photos can be taken with good lighting. Of course the quality is not nearly that of an expensive laser 3D scanner, but it’s getting ever cheaper and easier to do these things, and you only have to be creative in order to come up with new possibilities (interactive museum displays, reconstructions of destroyed sites, archaeological games using actual objects, VR in the classroom… you name it). My quest will be to test these methods and find applications for them within the field of Egyptology.
As another test case, I photographed the Hekate triformis with help from Marein Meijer at the National Museum of Antiquities. The lighting was awful, but this was one of the few statues that could be viewed from 360 degrees. This photoscan was done using 102 photos. There are some weird bits, especially on top, because the statue was quite tall and difficult to photograph. Also, lots of shadows. But here is the result:
Hi Nicky,
First of, great post! I wonder, though, whether these techniques (and which of them) result in true mesh shapes, such as that you get actual, deformable 3D representations of the object, or more like a 3D simulation, along the lines of what already used to be possible with QuickTime VR so many years ago (i.e. you stitch together a panomaric photograph of the object, wrap it around a cylinder, and as you rotate the cylinder, you simulate 3D, without it actually being 3D). Although both techniques have their certain advantages and disadvantages, hence uses, the QuickTime VR technique is more reasonably implementable than the creation of a true 3D mesh, yet delivers a way less versatile result. Any ideas on that?
Cheers,
Alexander.
Hi Alex, thank you for your reply! This technique creates a true mesh, see here: https://nickyvandebeek.com/wp/wp-content/uploads/mesh.jpg
I guess a mesh can be more easily integrated in 3D environments and edited using 3DS Max or other software, although the ‘sharpness’ of the object is not brilliant as you can see in the example. But unless you want to 3D print or render in super high quality, I guess these ‘easy 3D’ models can be quite useful for some purposes!