4 ms·
There's something I don't understand: I've heard that one of the keys for avoiding VR motion sickness is having both rotational and positional head tracking. A
by badsock 13y ago
There's something I don't understand: I've heard that one of the keys for avoiding VR motion sickness is having both rotational and positional head tracking.
A naive interpretation would be that for that to be possible from a prerecording, you'd have to have a 360 degree recording from the perspective of each cubic millimeter whithin the given volume of space that you'd expect someone's head to move.
Of course that's impossible, and there's certainly ways to interpolate from fewer viewpoints, but I've not heard of any that sounds like it's convincingly solved the problem. Is there one?
- tizzdogg 13y agoWithout even considering positional head tracking, just the act of rotating your head moves the location of your eyeballs in space (our eyes are not the center of rotation of our head, our neck is). So I'm also doubtful that this will end up looking any better than a simple 3d QuickTime VR sphere.
- MrScruff 13y agoHaving viewed one of these mono 360 clips on an Occulus I would say it's a significant step up from watching it on a monitor. However you're right that a truly compelling experience will require full 3D reconstruction of the environment so you can render both eyes in realtime with head tracking. That's probably feasible for some types of simple scenes but probably not a nature documentary.
- beekays 13y agoRegarding interpolation from fewer viewpoints, you might check out these demos if you haven't seen them. Requires Flash. http://lightfield.stanford.edu/lfs.html http://lightfield.stanford.edu/lfs.html I could certainly imagine a kind of 360 degree light field slideshow app with current technology. Not sure how much entertainment value is to be had there, though.
- badsock 13y agoThanks for the link! That's just what I was looking for re: interpolation. Looking their acquisition rig, however, it seems like a choice between a large array of cameras, or one camera and a still image. It will be interesting to see if that can be made to work as video, with a large FOV, and interpolated movement on 3 axes. Seems like a big challenge.