Week 8 3D1

This week’s post will focus on “photogrammetry” or the use of scanning objects using photos and turning it in to 3D objects. During class this week we were given the assignment to create a 3D mesh on top of a facial scan, the idea was to use a scan of our own faces. However the scan of my face was somewhat incomplete and lacked sufficient coverage for me to be able to patch together a model. I wish now that I had made a few screenshots of those incomplete models, because it looked really scary and deformed. So I ended up using the photo scan of a former student’s face that the teacher provided instead of my own face.

The main topology of the face is based around the mouth and eyes, with loops supporting the massive deformation that those parts of the face are capable of performing. The topology here is based on just that, which parts that move the most and which are more rigid. So looking at the anatomy of the face can be very helpful when designing the edge-flow for a face mesh.

It is really cool to work with a foundation that is as accurate as the scan is, where you are able to lift the exact proportions and unique features right off a face and into your mesh. It feels like this should be the future of game graphics, at least as a measure towards giving characters unique and believable faces.

I read the article on Photogrammetry at ‘Photogrammetry’ in ‘The Vanishing of Ethan Carter’ and was really impressed by the way they used the photo scans as a base for their specific art style. I also really liked how it discussed the visual complexity of nature and the difficulty of capturing that in game graphics. For me that complexity is to a high degree an issue of time restriction, I pay a lot of attention to the complex details that make up the whole impression of an object. But I often find it hard to justify the work hours needed to achieve those results against the timeframe of a game project, the fact that the time limit for the games we make at school is usually around eight weeks could factor in to this reasoning of mine…

The main potential with photogrammetry as I see it is to use the scans to provide the complexity of reality as a base and then use the artist’s own expression as an extension on top of that. That way you will have created an optimized version of modified reality, as the subconscious will probably buy the illusion of nature-made complexity in a setting that is completely spaced out and fictional. It is interesting that they mostly used this method as a way to break and rebuild the way they viewed reality, because that is the process that many artist go through in order to form new ways of expressing themselves through image distortion and abstract art styles. You have to understand the pieces that makes out the impression of an object before you can distort it, or make an abstract version of it.