h a l f b a k e r yThe mutter of invention.
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
There are personal LCD displays (in the shape of glasses) that simulate an image in the empty space behind them.
Works great, except that it makes some people motion sick, since the image motion follows the head motion rather than contradicting it, as would be normal for a stationary TV screen. (Reality:
you turn left, the TV moves right in your field of vision. Viewing glasses: you turn left, the TV moves way left.)
Now that very small tilt-sensors are here, maybe they could be built into the display unit with a simple, but fast image processor to make the simulated image appear stationary in spite of minor head movements.
Epinions on Sony's video headsets.
http://www.epinions...o-Headsets-All-Sony
"Got Dramamine?" [jutta, Oct 14 2000]
Body Position Garment
http://www.halfbake...0Position_20Garment
[Mickey the Fish, Oct 14 2000]
[link]
|
| |
VR headsets do this, of course, but (as you pointed out in response to the previous version of this annotation) they require cooperation from the rendering system. |
|
| |
I'm confused by your description of "exaggerated motion". The video goggles I've seen keep the screen absolutely centered as you move; they don't (can't) move it at all relative to your head. This is an issue, in fact; how would you move the screen to compensate, without mechanical contrivance? |
|
| |
Another issue is that a tilt sensor can't detect the most common type of head rotation (left and right), since there's no tilting. You'd need magnetos or a local positioning system or a gyroscope or something.| — | egnor,
Oct 14 2000, last modified Oct 15 2000 |
|
|
| |
Exaggerated motion: what I mean is that, as you would expect, the virtual TV appears as if it were mounted on a six foot stick coming out of your forehead. When you rotate your head only a little, the TV mounted at the end of the six foot stick moves a lot - the angle is the same, of course, but the distance from the viewer means that the screen jumps much for even minute movements. |
|
| |
Rotation: You're right; I've been glossing over the fact that tilt sensors can't detect rotation. I'm thinking that fixing one of these is better than nothing, but I don't really know how much better, or what to do about the rest.| — | jutta,
Oct 14 2000, last modified Oct 15 2000 |
|
|
| |
What you need are acceleration sensors: integrate these twice and you get absolute motion (including rotation if you use more than one). Acceleration sensors are common these days, and are widely used in airbag sensors. Due to advances in MEMS, they are getting much better, smaller, and more accurate. |
|
| |
Acceleration sensors don't work for low-frequency positioning. Integration leads to drift. Work out the error rates sometime. (They can be useful for high-frequency positioning when combined with a DC correction.) |
|
| |
The solution is obvious, though: you need to use the glasses lying down. That way tilt sensors can be used to detect head rotation, and it's more comfy besides. |
|
| |
You need the Body Position Garment, with hood attachment. This will reliably report the relative positions of head and members to within a gnats whisker. |
|
| |