Forum Discussion
Immersive 3D video query
Hello,
I'm a long time lurker, and have been using/testing Oculus since March (starting with DK1, and now DK2). I have a background in filmmaking and in 3D animation / compositing / effects work.
A thought came to me this morning, and I'm curious as to what the technical feasibility of this is. I imagined a pair of 2 UHD cameras (4k) capturing both color and z-depth information, and a program/processor to capture both images / z-depth information and putting them together in a virtualized / generated 3D environment for a user to walk around in.
The way I imagined this is to set up the 2 cameras far away from each other so that the foreground subject would ge 300+ degree coverage, and anything occluded by the foreground object would still be covered because of the 2 cameras.
The result, as I imagine it, is a semi low resolution 3D generated scene, playing back, based on the image and z-depth capture of the 2 cameras.
I've seen semblances of this created with those who are tinkering with the kinect devices. I also remember seeing the video playback of the guy playing guitar. But I'm wondering how feasible it would be to capture the live scene and digitize it in 3D so a user with a HMD could then walk around in that same scene. The resolution would be poor, ompared to the original 4k capture, but is it doable? Not a spherical video play. And not a 3D screen - but real immersive 3D where a user can walk around the 3D generated environment.
TL:DR - yes, it has major implications for the porn market. Or being virtually present in a place that has been captured this way.
Dillon
I'm a long time lurker, and have been using/testing Oculus since March (starting with DK1, and now DK2). I have a background in filmmaking and in 3D animation / compositing / effects work.
A thought came to me this morning, and I'm curious as to what the technical feasibility of this is. I imagined a pair of 2 UHD cameras (4k) capturing both color and z-depth information, and a program/processor to capture both images / z-depth information and putting them together in a virtualized / generated 3D environment for a user to walk around in.
The way I imagined this is to set up the 2 cameras far away from each other so that the foreground subject would ge 300+ degree coverage, and anything occluded by the foreground object would still be covered because of the 2 cameras.
The result, as I imagine it, is a semi low resolution 3D generated scene, playing back, based on the image and z-depth capture of the 2 cameras.
I've seen semblances of this created with those who are tinkering with the kinect devices. I also remember seeing the video playback of the guy playing guitar. But I'm wondering how feasible it would be to capture the live scene and digitize it in 3D so a user with a HMD could then walk around in that same scene. The resolution would be poor, ompared to the original 4k capture, but is it doable? Not a spherical video play. And not a 3D screen - but real immersive 3D where a user can walk around the 3D generated environment.
TL:DR - yes, it has major implications for the porn market. Or being virtually present in a place that has been captured this way.
Dillon
No RepliesBe the first to reply