"Holographic" Desk Allows Interaction With Virtual Objects

Zibodiz writes "The Sensors and Devices group at Microsoft Research has developed a new system called HoloDesk that allows users to pick up, move and even shoot virtual 3D objects. It's about the size of a filing cabinet and is made up of an overhead screen that projects a 2D image through a half-silvered beam splitter into a viewing area beneath. A Kinect camera keeps tabs on a user's hand position within the 3D virtual environment, a webcam tracks the user's face to help with placement accuracy, and custom algorithms bring everything together in (something very close to) real time."

Comments

Point Cloud ?

[UnknownSoldier]

I'm surprised they are using a point cloud and not a mesh for the tracking? That would explain why the physics seems a little unstable / jumpy when the ball floats around on the book.

Cool prototype -- will be real interesting to see what kind of applications get developed once this tech is cheap enough where every home has one.

Re:Point Cloud ?

[derGoldstein]

This is a shot in the dark, but: interacting meshes would involve specific collision detection between surfaces, which (when scanning an object in real time) could lead to locked models (the virtual object could "stick" to you, or other objects). A particle system allows them to define "particles repel each other" and get around that problem. Notice how (relatively) large the particles themselves are -- I'm guessing it can only deal with a limited amount of particles within the environment at the same time.
(and yes, I realize that the size of the particles you see doesn't necessarily mean that this is what the program is doing, but the number of particles probably is)

Already been done.... back in 98

[Anonymous Coward]

I worked for the Electronics Visualization Laboratory, back in the late 90's. We developed a similar system called PARIS,some time around 1998.

http://www.evl.uic.edu/core.php?mod=4&type=1&indi=83

Re:Now where did I leave that virtual stapler

[Amouth]

nooooooo don't let clippy out of there or we are all doomed

Re:HoloDesk?

[hawguy]

There's no force-feedback yet. You can interact with object in the environment, but you can't feel them. So this is of no interest to Picard, or the porn industry, for now.

Not on this one, but the University of Tokyo is coming closer to force feedback using ultrasound:

http://www.gizmag.com/tactile-holographic-display/12466/

Re:Sorry

[JMZero]

instead of relying on sensors that are orders of magnitude more sensitive or precise than the human hand

The hand isn't a sensor at all in this context; rather, the manipulations of the hand are picked up by distinctly non-hand sensors. It's clear the current limit on this system is those sensors, computers, and software - all of which could improve quickly if this moved beyond research.

extremely low resolution hands

My hands have, uh... pretty high resolution. Are your hands kind of blocky? Do they show aliasing when you turn them? More to the point, humans are extremely adept at doing fine manipulation with their hands and these manipulations are extremely intuitive. If they could make this work very well, I see no reason it couldn't be used for a bunch of things: teaching and demonstrating, as an intuitive UI for controlling robots (that might actually be acting on smaller or larger or toxic or distant objects in real life), or for experimenting with possible approaches or designs.

It doesn't take much imagination to come up with possible applications for this.

That said, probably it will never come to anything (at least not in a similar form) or not for a while. But if you disapprove because you fail to see practical applications now, I think you're both wrong (in this case) and misguided (in the general case). I think it's cool MS is doing research that they probably can't exploit immediately. It shows foresight to be thinking about interface methods before they're really practical. Nintendo (or whoever they bought tech from) probably had some very crappy Wii-like peripherals in research long before they worked well enough to sell. It might take 100 ideas and prototypes like this to find 1 that is the next big thing. But that's how we get cool new stuff.

Disregarding all that, even if it was completely pointless (and I don't think it is) I think it's fun that we can see it and discuss it on a site that is about interesting technology.

I mean short of being a toy, what is the point?

Do you mean "beyond being a toy"? If so, you said the exact opposite thing. Or do you not think this would fall short of working as a toy? Because it seems to me like it works as a toy right now.