Reality capture is such a cool concept—taking things from the real world and replicating it in virtual reality. It’s aspects like this that add to the immersion factor and blur the lines between virtual and reality. This concept reminded me a lot of the movie Inception. There was a line in there that read “Never recreate places from your memory…because building a dream from your memory is the easiest way to lose your grasp on what's real and what is a dream”. In the movie, losing your grasp on what’s real was a bad thing, but in VR it is precisely what we’re trying to do. 3D scanning software allows us to do that, and although the technology is not perfect right now, it shows a lot of promise of how we can easily scan and import real life objects into cyberspace in the near future.

I used two different techniques for reality capture, the Structure Sensor and 123D Catch. The first I tried was 123D Catch. I heard a lot of people were having trouble with this app. But it worked relatively fine for me. I scanned a vinyl toy figure of Scar from the Lion King. I did it in my apartment with artificial lighting and followed the references on the screen. I took a total of about 40 pictures. After a rather long processing time, I got a pretty decent model back. It had some holes in the mesh but overall I was pretty happy with it. I downloaded it from the 123D Catch website and imported it into unity. The model worked right away. I didn’t even have to use MeshLab to combine the textures.

The structure sensor was similar to 123D Catch but with a few key differences. The sensor is an add on camera instead of utilizing the iPad’s camera. The scanning is also using a video rather than multiple pictures. The results were okay after multiple attempts at scanning my head. I think each method has their own advantages. The structure sensor is good for scanning people’s faces while 123D Catch is good for scanning objects.