Compromises and designs

Looking through internet and books hasn't gotten me anywhere. There is virtually no information about 3D projection mapping, I assume because it's a rather new technique which has only recently surfaced. The only way I can discover more about 3D Projection mapping is by experimenting and learning for myself.

Of the information I've combed through and found, one thing strikes me as very interesting.

http://info.ee.surrey.ac.uk/Personal/Z.Kalal/

Recently, a researcher in computer vision at the University of Surrey has come up with an excellent method for tracking objects via a webcam. I could potentially use this to track my body parts movements and overlay a 3D character over my actual body via projector. This would eliminate the need for tedious trial and error training in keeping in sync. I could even have audience participation by having it track symbols on an outfit. Unfortunately the source code hasn't yet been released and likely will not be until way after my deadline. Even so, I may revisit this project in the future and use some of Zdenek Kalal's work. 

The video above shows me at the end of an extensive trial and error run on trying to keep in sync with an animation, which was based upon a previous recording of myself in front the grid I previous made. After the initial video was recorded, I distorted it to overlap the original grid so I could reference it and trace the shadows and create an animation which would be projected onto me, if I were able to keep in sync.

From this test I've come to the conclusion that projection onto a live actor is something which is impossible for me at the moment. The light is blinding so it becomes very difficult to see where I am in relation to the character by looking at the screen, and by looking at the shadow I still can not keep in sync properly. Maybe if I had a year to perfect this it would be possible and amazing, but for now I should simplify my project so that the final piece will look more professional and polished. Plus, using only still props means that I can reproduce this performance with ease, and many more people will be able to see it, instead of just the few people who can see the one performance.

These are just some rough sketches I made for character and set design. Initially I wanted to explore sets which were obviously reminiscent of traditional theatre sets.  Particularly I wanted to use a thrust type of staging, which is the classic theatre set everyone should be familiar with, where a platform comes out from the background. I wanted to use the thrust staging as it linked the animation with traditional media and entertainment, showing the progress we have made since then and showing how everything can be emulated, which I want to be a big part of my animation. Unfortunately, the area where my animation will be projected is unsuitable for this and will look rather silly, as the set will be small and high up due to the limitations of the placement of the projector.

I've decided that I don't want any particular character models, as I want the two character to be able to morph into whatever fits their mood, showing the fact that people can easily change what they look like as they can change their avatar in an online game. Courage the Cowardly dog has inspired me somewhat with this as he changes forms to communicate and it's that cartoony, energetic feel I want to capture in my animation. Using paper and paint, I've filled in the eyes in two white masks I've had, as well as going over them again in white paint to ensure they reflect the light better. By filling in the eyes I can project my own eyes onto the masks and have a higher level of control over the emotion shown in them. The only problem I'm worried about is the texture I have after the painting, where before they were smooth, they now have thin lines all over them and are very hard, although I don't expect anyone to be touching them.

In the set design I've tried to keep in mind how the light will bounce off the set. The angles of the objects are intentionally set so that shadows don't become a hindrance, as anywhere there are shadows, I can't project, thus can't manipulate. The other thing I have to be aware of is portability of the props. I need to be able to transport the entire set to and from my house and to the University where it will be shown, and since the projection has to be so specifically set, it can't be too complex to fix in specific places, perhaps I should create a mechanism to ensure props stay in perfectly the same place. Because of this, and various other issues I should have already mentioned, I have decided to keep the set as simple as possible, and this also gives a minimalistic, futuristic, "apple" type of feel to it. This will also be an excellent contrast to the colourful, exciting animation being projected onto it.

The masks should be slightly off centre, and not facing head on to the audience to create a more organic feel to show that the masks are alive and make them seem more human. There will also be a board, I have recently painted white, just for an extra, simple prop which will let me show off projection mapping a bit more. 

Over the next few days I will be looking into getting feedback from peers in order to improve my animation. I need to do this soon as I will be starting the animation as soon as I finalize the story, storyboards and create a map of the are where my animation will be projected.