Project 3 :: Caitlin Boyle & Asa Foster :: We Be Monsters
Asa and I were clear and focused from the very beginning: from the moment we got our Kinect, we were hoping to make a puppet. We wound up with something that can be controlled with our bodies, which is a step in the right direction, but I think we hold the most stock in the process that got us there and what we can do to improve our project.
The first major problem was my laptop… which refused to install OSCeleton, which did all of the talking in between Processing and OpenNI. Because of this itsy bitsy snafu, all debugging and the majority of programming had to be done on Asa’s computer, which was only possible when both of us were free. Despite this, once we got things running on Asa’s laptop(which didn’t happen until Wednesday of last week) we took off running, using primarily a Processing sketch by OSCeleton’s creator Sensebloom, Stickmanetic. Our original plan was to create a series of puppets that could be controlled by two or more users, using Kitchen Budapest’s Animata software, but we quickly hit our second wall: Animata can only take in one set of skeleton points at a time, as it uses a limited OSC mode that does not send user# information. We could not use Animata, as we had planned, to make cooperative puppets: after trying and failing to get max.msp.jitter working with .pngs (max isn’t really an image-friendly software, it much prefers sound and video), we decided to bite the bullet and try to re-create a very basic Animata in Processing.
We wanted to make puppets that would be interesting to interact with, and that did not adhere to human anatomy; I sketched up a Behemoth, a Leviathan, and a dragon and handed them over to Asa to get cleaned up and separated into puppet parts in Flash.
We brought the .pngs into Processing, got Processing to recognize two separate users and assign them puppet parts, and we got the puppet parts following our skeletons, but then ran into a problem that took up the rest of our night, and ultimately spelled our defeat: ROTATE. We were trying to link the .pngs with pivot points that the pieces would rotate around as if they were riveted down, but no matter what we tried we could not get the rotating pieces to behave correctly. In the interest of keeping the puppet from flying off into the abyss, we scrapped the rotate function-for now- and made a much stiffer puppet that sits directly on our skeletons, rather than a puppet that is controlled by our skeleton but keeps it’s own, non-humanoid skeleton. (Click any of the following images to view a video of the Behemoth doin’ it’s thing).
Behemoth’s got chickin’ legs (for now- we drew bones from the hip to the knee on each leg as a temporary fix for Wandering Foot Syndrome).
Asa and I trying to work out how to walk forward, back to back (documentation of our physical puppetry process on it’s way).
DESTROY THE COUNTRYSIDE.
It is incredibly difficult to control coherently- it takes a lot of back and forth conversation between the front and back half of the behemoth to get anything that looks like a solid creature; in the featured video, I am controlling the front and Asa has the back. It’s also challenging to move your body the way the puppet needs to be moved, but I think this works FOR the puppet… in order to be the puppeteer, you have to learn how to move your limbs in counter intuitive ways.
We re-wrote our code, and got our Behemoth off the ground!
Hi Asa, Caitlin. Great work. Nail the frame rate issue, then make a dope video. Here’s the feedback from the crit pads:
————————————-
Great presentation — nice overview of the problem, and nice visual research (Chinese costumes, Snuffy, etc.) The idea of a multi-person puppet is very good. It must have been difficult to … test! The goal was ambitious, the results are rough but very charming, and you learned a lot.
Very nice sketches – I already clearly understand what you are planning. Nice presentation in general. Tech issues! But we got through to to demo. This is a cute project, I think you should try buckling down and porting it to OF to speed it up (Or get a better algorithm as you said).
I have to say though, I’m NOT digging the self-deprecation. If you don’t know how to write code well, that’s fine, but calling attention to it again and again doesn’t need to be mentioned in your presentation. If you’re confident enough to show something, be confident in it – don’t justify or qualify it with your own skill – that’s just bad form.
Those are freaking awesome drawings. Beautiful monsters and very easy to understand. Terrifying yet adorable. I would of loved to see all 3 of them. You’ve got a great base to work off of. I’ve got full confindence you can get this up and running.
Great variation on the puppet idea – I agree the human puppets can get old.
Also, very interesting insight into the puppet world – multi-person puppets are definitely NOT the first things you think of. Agree!
Very nice concept art. You presented a clear background and precedent and talked about what you were doing differently. Thanks! i like how it forces the puppeteers to lean together. Frame-rate is disapointingly slow. Any plans for 3d in the future?
Nice idea. Maybe some more subtle mapping from limbs to puppet?
Love your project. Good job!
Awesome looking monster. Really fun concept too. If you’re willing to still tackle the technical headache, it’d be cool to expand on this (your drawings of a dragon controlled by four people would be pretty epic to pull off).
Good intro already. I like how you switch off whos talking between slides.Great concept sketches. Great connection flowchart, lets us know whats going on as well as the decisons you made with software. Needs a fps increase for sure.
nice looking slides and sketches. I kind of like how its slow, it looks cartoony.
Really nice introduction. Like your sketches. Your monster looks super cool! I like the movements you chose to control the different parts of the monster. Really fun concept to have two people controlling. It is clear that a lot of effort went into this project.
Good slides and explanation. Nice slide titles and dry humor
Nice puppet.
Cute idea, really forces interaction between people.
Great interaction with multiple people. I imagine that was tricky but has great potential!
The video in the presentation (when the monster first moved) was a cool touch, nice.
Great project – who contributed what? I really like this concept and your artwork is gorgeous. I think you should be able to make it go much faster, but it sounds like you know that.
Ha! cool project. I like the thing with two people interacting. That is really fun.
Love that you guys get to lay claim to the first multi-person Kinect puppet, right on.