/diy/ - Do It Yourself

>>599273
If they are lined up perfectly or you can align them perfectly via software then yes. You'll need to play with the distance between them a little. There's a weird thing that happens when they are too close together and not far enough apart. Part of that has to do with the distance the landscape is around the cameras. So, you should have a way to change the distance between the cameras. Even better if it can be done easily on the fly. Think of it like a manual focus that you have to get used to doing to make it look right in the final product..

i highly doubt the signal you get out of a camera, and the signal that drives the Occulus Rift are compatible.

>>599273
The distance between them is dependent on the focal length of the lens.

Stop giving military contractors like DARPA good ideas. Seriously though there's nothing stopping this in the future; Remote soldiers using live feed stereoscopic cameras.

>>599330
That was actually my intention

I've got a project coming up for my EE degree and I was just mulling this idea over in my head for a while

Why not mount them like, a foot to either side and give yourself super depth perception?

>>599340
Why not a kilometre apart?

>>599333

You're gonna get bagged by men and suits yo

>>599340
>>599342


Holy fuck yes this would literally make you feel like a giant if you put it up in the sky.

>>599342
>>599344
>yfw they already do this on a much bigger scale with telescopes in space

>>599345

I always knew this but this sort of thing has never been attempted for recreational purposes. Imagine two cameras in New York place on nearby skyscrapers looking down on the city. Charging people 10 bucks to see what it looks like to be a giant. It would be cooler if it moved along with your head movements but that would be harder to pull off.

>>599347
Actually OP here, that gives me an idea.

I was originally going to put the optics into a totally transparent RC cockpit and see if it would feel like a bird, but it turns out it would be way too front heavy for the RC plane.

I'll go and talk to the dude who owns the plane and see if we could put them on the ends of the wings. It'll probably be much more easily pulled off.

>>599349

Get two lightweight laptop cams, use those.

>>599273 (OP)
Already been done there was a crowd sourcing project shits never gonna happen he is asking for too much
http://www.indiegogo.com/projects/3d-webcam?c=gallery

fyi on the Oculus Rift, the separate left and right images are displayed side-by-side on a single screen, not separate screens. You'll need a way to combine the images into a single video feed. It also needs to be pre-distorted to correct for the distortion caused by the lenses. Take a look at some Youtube videos, and you'll see some video captures that show what the video going to the device really looks like. Or see picrelated.

>>599411
Ah I see

That does sound a lot more difficult now

>>599416
Look up OpenCV, it would be pretty handy for that type of on-the-fly video manipulation

>>599330
you are a faggot, guess why Lunokhod has 2 cameras
if you think these things weren't done since decades then you are an idiot

>>599416

Manipulating two images on one screen is magnitudes easier than manipulating two images on two cameras (especially if they're not properly aligned).

>>599282
Me here.

This is what I was referring to:

http://en.wikipedia.org/wiki/Stereoscopy#Longer_base_line_for_distant_objects_.22Hyper_Stereo.22

>>599411
>>599416
But there's already programs you can do that on the fly with. It is no more difficult than learning how to use MSPaint, if that hard.

>>599340
why not just mount a range finder on it and feed that to the oculus rift

>>599416
This is not a hard problem, but it's a software problem so I'm not sure if it would apply for an EE project.
All you need is a distortion model for the cameras and a distortion model for the occulus rift and write a shader/program to convert between them.