Poor Man's Stereoscopic Projection

Jed Link writes: "This summer I helped built a Geowall stereoscopic projection (3D) system for the Southern California Earthquake Center. Although there are no new concepts involved with this system, what is new is that the system cost a little over $10K and is comprised of materials that you can buy at any computer-hardware store. A complete description of the system, as well as a diary of its assembly is available here. Traditional stereoscopic projection systems like The Cave which is used primarily for new product modeling and on a few university campuses cost anywhere from $150K to $1.5M. They are built into a fixed location, often requiring significant architectural modifications, so transportability isn't even an option. The Geowall, on the other hand, can be fixed to a cart (like we've done) and rolled from room to room. The price-tag makes the system feasible for undergraduate post-secondary education classrooms and even high schools. The system is based on a very simple concept, and while its use is currently primarily educational, I think it's only a matter of time before we see something like this in the gaming or entertainment industries."

Comments

$10K?

[Anonymous Coward]

Not exactly poor man then. More like upper middle class.

Oh well, back to looking at my 14" screen.

Sure...

[Quixote]

It costs 10K now, but once the p0rn industry gets into the market, we'll all, errr.. I mean, you'll all be able to buy it for under $1K with economies of scale.

Missing the point

[Anonymous Coward]

The author is completely off the mark in comparing his "geowall" ($12k, mobile, stock parts) to a cave system ($120k, fixed, custom parts). The projectors are easy for a cave, too. They cost a few thousand dollars, and use the same NVIDIA trick as he does. What is expensive and difficult is the motion/orientation-tracking system. The last cave I was in used a magnetic field to precisely detect the position and orientation of the viewer, thus moving the image ot make it seem like you were crawling around inside the scene. This has to be pretty precise, IIR, or else it feels wrong. And getting a good rectilinear magnetic field the size of a room takes lots of metal -- hundreds of pounds of shielding, permanent magnets, electromagnets, etc -- positioned very carefully all over the room. Not to mention the sensors. This is the part that costs $100k.

Re:Missing the point

[cyranose]

I don't know about the CAVE you were in. But on the ones I've built and/or been in, the trackers weren't necessarily the driving cost. Here are some of the big costs in the past:

1. 4-6 high-end stereo-capable video channels (typically big iron SGI or smaller but heavily synchronized computers) doing 120hz rendering of large datasets

2. Infrastructure -- custom-built cube, often raised off the floor for bottom to work, requiring risers, stairs, and quality construction for insurance purposes

3. 4-6 x 8'-10' fixed panel rear-projection screens (that preserve polizaration, not an easy trick)

We were able to reduce #2 with wood construction, omitting the raised floor. We also found that for entertainment apps, head tracking wasn't the most important thing, so for #1, somewhat cheaper computers and no tracking is an option. For #3, cheaper acryllic screens are fine if you don't need passive stereo (which you don't if your virtual objects are greater than say 12 feet from the viewer or you use active stereo).

Bottom line: we built a cheap-ass CAVE in 1993 for under $30k not counting the computers (which were over $300k back then). With new PCs, new projectors, and some clever predictive synchronization over simple ethernet, we could be talking $6k per wall without stereo, $10k with. Add decent tracking, and you add maybe another $10k overall. Add labor costs, markup, and an insurable level of steel infrastructure and that gets you near the $120k for current CAVE systems. But cheaper is always an option. For example, in my home, I use non-stero and an 8' screen made of stretching a 300 thread king-sized bed-sheet over a custom wood frame (it's as good or better than many screens I've used).

A thesis work related to the subject

[jukal]

I read this Licentiate Thesis work sometime ago, and if you are interested in getting to know virtual reality/environment techniques and CAVE construction it is excellent. It's PDF and over 700K, 146 pages.

SGI

[lovebyte]

These things are quite commonly build out of one expensive projector and an SGI box. There are many of these and they are mostly used for viewing 3D models (of geological surveys, cars, proteins, ...) for a group of people. If you are alone, then you can easily use active stereo on a big CRT screen.
I think it is interesting that these people have done a cheap version of a one-wall cave, but caves are most impressive when you have more than one wall! I've personnaly used caves of 3 (floor and corner) and four walls (floor and 3 vertical walls, i.e. a room without ceiling and one wall), and I know of people who have played Doom in them. They told me that it was physically tiring because you had to move, duck, ... Maybe that would be a good training for geeks wanting to work for the FBI!

Didn't someone do this in 1996?

[reachinmark]

Actually, I know they did: The Wedge. A much more immersive version of it too. Granted, the wedge uses the expensive Crystal Eyes glasses for stereo instead of polarised glasses/projectors, but the basic principle of cheap VR is the same. Why don't people who set out to make a cheaper better solution start by doing a bit of web surfing!?

Is it useful though?

[bluGill]

Sure this is cool, but is it useful? 2D screens are old hat, and seem to work well. People play doom and other 3d games on them without problems, so we can fake the 3rd dimention if we try.

I can't be the only person who gets sick watching 3d films. I've only done it a few times, but that is enough that I refuse to considering doing it again. If 3d films become a major part of education, then I'm disabled because I cannot watch them.

Try it with NO hardware... just crossed eyes

[Goldenhawk]

No kidding. Start with this picture (JPEG, 144Kb) of me on vacation last year. Now position your head squarely in front of the monitor, about 24" from the screen, and cross your eyes slowly until you see three pictures - the center one will be in stereo.

This is rather like the random-dot stereograms, but inverted left/right from that arrangement. In the RDS, you RELAX your eyes, the opposite of crossing them. I personally find this difficult, so I swap the images so crossed eyes produce the correct left/right arrangement instead.

Incidentally, I used to fly Microsoft Flight Simulator back in 1988 this way - yes, version 1.0. I discovered that I could set two different forward views from the Chase Plane mode, with one plane offset slightly to the right and the other slightly to the left. By properly arranging the windows and crossing my eyes, I could fly around looking at the simulated world in true 3D. I believe you can still do this with the different window options available in Flight Sim, and you could probably do this in any game that allows you to set up multiple windows from different viewpoints.

Now, granted, it is tough on your eyes, and it's kind of hard to see any non-stereo items (like the control panel), but it IS 3D and requires NO hardware. From time to time I do this for other purposes - like the picture above.

You can also do this with any camera if you have a still (or mostly still) subject - take a photo, move sideways about four inches and take another. Then load both photos into your image editor of choice, position them side-by-side with the proper left/right orientation, and you're set.