Headtracking projector

[cadcoke5]

I just came across this interesting combination of a projector and head tracking software.

http://www.picoprojector-info.com/immer ... -projector" onclick="window.open(this.href);return false;

It has a camera and software tracking the head motion of the user, and then a device moves a small projector so that its image is projected towards the users direction of view. The image is projected onto a cylindrical screen sitting on a desktop, with the projector sitting in the center.

The demonstration has two major limitations. #1) There is a slow response speed. #2) The projected image is very small. However, both of the above issues should be easily addressable. E.g They can switch to a different head tracking technology with less lag. Also, they don't need to use a pico-projector, but can switch to a larger one with a wider area of coverage. If the weight of the projector is too much to move quickly, they can aim the projector at a mirror and then move the mirror which has a much less inertia.

Also, for you home-brew folks, I think even the head tracker can be eliminated by having the projector mounted in a frame that is situated over the user's head and then the movement of the projector tied to the swivel of the chair. Someone here actually built this, but I can't find the discussion for some reason. I am not choosing the correct search terms. It was discussed within the past year.

An obvious alternative is the multi-projector camera set-up used in cave environments or on a larger cylindrical screen. Also, just using a single projector and projecting onto a cylindrical mirror onto a cylindrical screen gives a wide field of view. I imagine the advantage to having the projected image move is that you can use a single projector and still obtain a high resolution image regardless of the direction you look. With falling projector prices for high definition it may not be an issue any more. It may be easier in the long run to use multiple projectors on a curved screen.

I will mention one more concept, which is s bit off topic. I didn't think it was worth a new thread.

For these types of displays, you always need to have a pop-up window that shows your controls or other game information. If this information were instead put on to a LCD panel situated about where a laptop screen would be, then you can have your flight simulator instrument panel and other similar displays on that panel. With a touch screen, I imagine it would be very useful. This idea it so obvious, I am sure it is my inexperience with gaming software that explains why I haven't seen this before. Can someone mention a game that supports this arrangement? E.g. a laptop with instrument panel displayed, and using a larger separate monitor for the display of the environment.

Joe Dunfee

[cybereality]

I saw that video too, but it seems useless to me. With such a nice curved screen, why not go for triple-head projection ala Nvidia 3D Vision Surround? With that small projection they are using I'd almost rather just look at my phone.

[Johnny-Mnemonic]

I think much better combination of Projector + Head tracking, for home VR, it's one-wall-cave.
Check this out:
http://www.vrgeeks.org/vr-kits/wall

[cybereality]

I think much better combination of Projector + Head tracking, for home VR, it's one-wall-cave.
Check this out:
http://www.vrgeeks.org/vr-kits/wall

Now that is actually something that looks fun and useful. Unlike the original video, which is fail.

[cadcoke5]

It is interesting that you say both fun and useful. I have yet to see VR really make any substantial progress towards being useful. Yes, there are demonstrations to show how things MIGHT be useful. But, they don't actually seem to sell many VR systems or get software vendors motivated to make VR compatible software. The demonstration head tracker that I posted about in the first message is a good example. It was a bit interesting to see it being demonstrated, but in the end, completely useless.

I have worked in the field of Computer Aided Drafting and Design for over 20 years. And even though the hardware to do a "one wall CAVE" type of VR is very easily affordable to any business, and has been for a decade, businesses don't use it. CAD software companies that used to have a switch to turn on stereoscopic display on the video card 10 years ago have actually taken that functionality away because no one used it.

Perhaps I am being too strict in my definition of VR. Perhaps the real-time rendering of an object in my CAD program as I move my mouse to rotate the object is VR. Well, at least it seems to be VR enough that people aren't motivated to purchase a system with head tracking.

Joe Dunfee

[cybereality]

Well I meant "useful" in that it serves a purpose. That purpose could be entertainment, novelty, etc. And that its actually providing a functional experience.

[cadcoke5]

I agree that entertainment is a useful function. If my next job (I have been unemployed for a while) lets me play with a projector, I will probably get the rest of the gear together to make a "One wall CAVE" just like in the video above. It may be a pipe dream, but even just playing with the possibilities of VR is a fun thing to do. Perhaps all the non-gaming VR guys are, in reality, role-playing a game of Star Trek. They are pretending they have a functional holodeck.

Joe Dunfee

[cadcoke5]

Just a bit of an update. The makers of the head-tracking projector I originally posted about have put up a more in-depth video, and show it playing Batman Arkham Asylum. They comment that it is easy to interface their project to most games.

http://www.youtube.com/watch?v=iV2oKaQA ... er&list=UL" onclick="window.open(this.href);return false;

The project was by students at;

University of Texas at Austin
"Motorized Motion-Controlled Display"
Christopher Koelling, Edward Wang, and Sarathi Sathasivan
Supervised by Tejas Ganapeth

They don't seem to be taking the project any further at this point. I did an internet search, as well as a search on the university sight, and there wasn't any more information.

Joe Dunfee