Procams 2006

Procams 2006 (see previous post for explanation of event) went very well today. I met some very interesting people and got a lot of positive feedback on the Wiremap. It was also great to talk continuously about the project, and seeing which questions came up most often. The questions most commonly questioned was how long does it take to calibrate, what is the best angle to view it at, how was the wire placement designed, and what were the wires made of.

The Wiremap takes about 30 minutes to calibrate. I use a calibration tool, and then I run the calibration program for them, where you line up your projector in such a way that blue is the dimension closer to the projector, and red is further away. A video of this is available on YouTube here.

Because the Wiremap is truly volumetric, it should be consistent from any point of view. It was funny to get this question over and over, and it sort of pointed at the idea that in order to engage with certain technologies, one must place oneself in a proper, conventional relative position.

In terms of wire placement, each wire must correspond to a sliver of light, but its depth is independent of that sliver (how far or close it is to the projector). The wires are placed randomly across that independent dimension.

I did at one point consider making a pattern with the depth, but it turns out there are quite a few advantages to randomizing it. When random, the overall aesthetic hints to the fact that there is no optimal point to view it from, while if it were patterned, certain points of view would be better than other points of view. I decided to steer away from this because in nature, when we interact with real 3d objects, we never worry about looking at it correctly - it's either there or it isn't.

The last question, what were the wires made of, came from the confusion that many people had that the wires themselves were emitting light. They don't. The wires are glossy white, so they reflect light really well. The wires were made of cut and straightened coat hangers, and the projected image is the only thing that creates the 3d model.

I met some very cool people... in no particular order:

• I met Jason Eppink, a friend I met on YouTube, for the first time in person. He had a lot of insightful aesthetic observations
• Jeff Han, the designer of the Holodust, the inspiration behind the Wiremap, was there to talk to me. He was surprised at how convincing the whole thing was. He also remarked that the display was way more convincing when dynamically moving.
• Perry Hoberman and Mark Bolas, USC faculty and colleagues were there. They seem to have been doing quite some work in this field using projectors to build volumetric renderers. They had a lot of insight, intriguing questions, and quite a bit of experience with the principles behind the project.
• Laura Venner a, "NASA/JPL Solar System Ambassador," came by and whipped out her hunk of aerogel (a substance that is 99.8% air, .2% of silicon dioxide) at the request of the guys from USC and shot some projector beam through it. Awesome...

I also wrote up an artist statement and technical information packet for the event, which can be seen here.

Most people understood the globe right away, and most people had difficulties wrapping their minds around the cube. Some people couldn't get the globe until after it was explained to them, and a few could comprehend everything about the cube in all rotations.

Thanks for everybody who showed up. Keep checking back as I embark on making either a 6 foot tall one, or one that has 256 wires (or a 6 foot tall one with 256 wires).

There were alot of photos and video taken at the event. If you happen to be one of those folks, I'd love to get copies of those pictures / videos. My email is hwang.al@gmail.com.

Wow, I'm becoming one of those bloggers who has way too much disorganized content... a textbook blogger. I'll work on it, I swear.

Wiremap is Mouse-Controlled

Here's another video of the Wiremap. Not too much new here, except its X and Z axes are defined by the mouse, rather than a keyboard. The analog feel really frees it up, although it still feels constrained to 2 dimensions. I'm struggling to get the P5 glove to work with the Wiremap. My comp sci roomie said he'd be able to help me out with calling the P5 drivers from Java. If I could get that to work, then I'd be able to make the motion along x, y, and z feel really seamless.

Here's an applet of the program that runs it:

http://homepages.nyu.edu/~aph220/mouse_controlled/

Why Sinatra? I randomized iTunes and this one was the first song that sounded appropriate.

I also managed to get a cube to float and rotate. This was a problem, however, because there was way too much information for a human eye to take in. So I think I'm going to abandon the cube until I get a Wiremap with 256 wires (perfect for a screen at 1024 x 768).

Also, a friend saw the Wiremap last night for the first time and said he didn't know how big it was. So just to give you an idea, it's about 25 inches wide, 16 inches tall, and 20 inches deep.

Another friend of mine was wondering if she could look at the submission I wrote for Procams, so I thought I'd share that on the net as well:

http://homepages.nyu.edu/~aph220/Wiremap.doc

I updated my portfolio entry for Wiremap with interesting observations on some of the differences between surface 3d rendering and volumetric 3d rendering. Check it out.

Procams

About two weeks ago I sent in the Wiremap an art submission to Procams2006, an event that focuses on projector and camera art. I didn't know whether my submission was quite appropriate because it has nothing to do with cameras, but then again, the public call for artwork says that they're looking for works that (among other things):

• uses active and non-traditional projection techniques
• involves projection onto custom screens, surfaces and objects
• deals with the aesthetics of projected light and digital projection

I'd been anxiously waiting to hear back from them ever since sending in my submission. I got this email today:

Dear Albert,

We are pleased to inform you that "WireMap" has been accepted for the ProCams art exhibit, Projecting Off the Wall. We loved the work and look forward to meeting you.

The exhibit date is June 18th. Our schedule is:

8-noon - setup

1pm-6pm - exhibit

!!!!!

Awesome. It'll be on the fourth floor of 721 Broadway, just one floor above where most of my undergrad theater took place.

I may have to temporarily abandon the production of my cube and really spruce up what I've got to make it as 1337 as possible. A portfolio with ideas of where this project is headed, ideas on how to integrate it with Jitter, possibilities with Maya plugins, with the P5 glove. I'd want to clean up the code a bit to make it more readable, too, just in case anybody asks.

So in a nutshell:

Sunday, June 18th, 1-6pm, 721 Broadway.

If you're in NYC, come!

Cube and P5

I got a P5 in the mail yesterday. Awesome.

It's cool, but it isn't as seamless as I had hoped. I feel as clumsy with my P5 as my parents must feel when they started moving that mouse around for the first time. The width and height dimensions track pretty well, which is what the glove uses to emulate a mouse.

I played one of the games that came with it and soon saw how twitchy the depth dimension is. I'm not sure why, but it will flip around to way far away to way close. At any rate, I've found a hobbyist who's developed another driver that apparently works better than the manufacturer's one. I'll try to install that tonight.

In other news, development for the cube is going well. Here's a little preview of the image with its edges.

I might be able to build it by this weekend, but then again, I might be heading over to DEMF. I can sharpen up my tuts and clean up on my digitz for only $40 for the entire weekend.

Obsessively Programming

I was buying tickets on Fandango last night for four people. They have a pulldown for how many children, adults, and seniors you would like to buy tickets for - the default is zero. I almost selected "3" thinking it meant four, because computer programs start counting at zero instead of one. That would have been really embarrassing.

"Oh, sorry guys. I accidentally bought three tickets instead of four because I counted myself as zero."

I've been obsessing over this Wiremap maybe a little too much lately. Here are the discoveries I've made for myself:

Because I come from a digital background rather than a physics or architecture background, the Y axis represents up and down, NOT forwards and backwards.

The Z axis represents forwards and backwards. And guess which one is positive? Forward is obviously positive.

This puts me at ends with the classic physics and architecture models. For example, as a grad physics student friend of mine noted, under my model, the charge of an electron is positive, and the classic "right hand rule" for physics equations all become left hand rules.

I've done lots and lots of consideration - probably a bit too much - on this point.

At any rate, I now must move on. I'm in the process of building 3d cubes and such in my Wiremap. I hope to be done in a week.

Wiremap Works!

And here's the program that runs the thing. (And if you missed it, here's the video explaining the project.)

So it's finally a relief to see exactly what I was envisioning when I came up with this project. It feels very 3d (even more so in person where you have more depth perception than from a single camera), and is very alive. It really feels like cyberspace is finally coming out of its box to really play.

The guy who owns the projector is a theatrical multimedia designer just starting to get into Jitter. We're thinking that if I can open up some input sockets in this program, he might be able to control the thing from a music visualizer. L337.

In the next couple of weeks, I want to really clean up the code. I also would like to build a few more programs - a bouncing ball, a rotating cube, and other things with elementary and natural motion, like collisions and gliding and hovering.

I also would like to talk with people who are all into the Maya and AutoCAD thing, because there has to be a way to configure a plugin, right?

On top of this, I'm still waiting on my P5 glove. If all goes well, with the glove, I'll be able to digitize a human hand into pure, digital light. Then interactions with digital 3d.