Slashdot Mirror
New "Power Glove" for the PC
friedmud writes "I just saw an interesting technology demo over at Essential Reality. It is of a "glove-like" device that allows you to interact with your computer. In particular the demo(which is very viewable using the CrossOver plugin) having to do with 3D Studio Max was just incredible (almost too incredible - but hey, we'll have to see) - they put together a model of a mouse in seconds by "molding" it using the "glove". It looks like the SDK is out, and the product is supposed to ship soon. News blurbs can be found at: PCWorld, Yahoo"
That demo looks totally fake. It looks like they recorded a video of someone playing Quake, Half-Life, and modeling in 3DS Max with a mouse, then superimposed that picture of someone acting out the movements. I would be very surprised if those on-screen movements were actually made with the glove in real-time. Besides, with no tactile feedback, I imagine it would be pretty hard to "grasp" objects.
main(c,r){for(r=32;r;) printf(++c>31?c=!r--,"\n":c<r?" ":~c&r?" `":" #");}
A glove interacting with a computer has quite a long way to go before it will be able to compete with the humble wad of clay.
The kinds of tools available to a sculptor, the tactile feedback of the medium is far, far superior to a single glove acting on air and the artist looking at a computer screen.
The glove has it's uses: it is not a replacement for a 3D capture device, however.
Potato chips are a by-yourself food.
What if it were not intended to replace clay, but use a combination of clay and the gloves to bypass the scanning step? The sculptor could put on the gloves, start working with some standard lump of clay, initialize the glove tracking program, and manipulate the 3D model with the gloves as their actual hands manipulate the real clay model. When the real-world sculpture is finished, the 3D model would be done, as well.
Then all that's left is to hook up the fabricator, hit "copy", and start cranking out toys...
hmm that would be a good idea, but you'd have to have EXTREMELY high-accuracy position tracking, otherwise the minor errors would quickly add up and you'd be left with a 3d model that looked absolutely nothing at all like the piece of clay. This is the same as when a beginner writes a 3d program which uses the standard SIN/COS rotation formula on the verticies that have already been rotated; everything looks ok to begin with, but soon the model loses all touch.
so, the 3d scanning method, or the completely virtual clay methods are the only ones that are at all feasable unless the detection accuracy of the gloves is insanely high.
one day....
ìì!
There's another potential problem with this: sculptors are unlikely to use just their hands to shape the clay. To bypass the scanning step, the glove tracking details would have to be combined with detailed information about every tool used. For example, was the tool that was dragged across the surface of the clay scraping with a flat or curved edge, or was it gouging a thin or wide groove?
An interesting idea, but I don't think the scanning process will be retired any time soon.