Are you ready for the newest technology that will be indistinguishable from magic?

Very neat! Do you know the file size and resolution? Are we talking a file that is only slightly larger than the standard bit maps or today or does it require something an order of magnitude larger?

Or… Is it a stereo camera with excellent post processing?

Wow that is amazing and horrifying all at the same time…

I’ve been thinking about this. Here’s how I would do it. I start with a two lens stereo camera with wide depth of field lenses. I would post process the picture and make transparency and focus a function of the image separation.
Before processing each image is in focus – or as “In focus” as a typical short focal length, higher f stop lens. It is the post processing that gives you the “out of focus” parts.

If I guessed correctly file size should be only ~2x the typical file. Very well done! I like it.

I can’t wait for the video camera version.

I wonder if it is possible to form a “composite” image from a lytro image, were every point is in focus all of the time. This is how someone might be able to do it:

1) Trace the rays of light back to the points on the object. This is creating a 3D image.
2) Create an observation point and a veiwing direction.
3) Create a 2d plane perpendicular to your veiwing direction, in front of the observation point. This will become the “canvas” for the final photograph. The distance between the observation point and the plane is the “viewing distance”.
4) Trace new lines from all of the points of the photograph, through the “canvas”, to the observation point.
5) Where a line crosses the “canvas” will be where that point is on the composite photograph. The canvas becomes the composite image.
6) When you stand in front of the photograph the same distance of the viewing distance and in the right place, the photograph will have the perfect perspective, as if you were looking from that point in the same direction where the lytro image was taken.

I wonder how this will affect the world of artificial intelligence. It is very hard to give a robot “eyes” and have it tell where the boundaries of objects are. But with a video camera form of lytro will allow computers to “see” 3d images without the need to put two images together in binocular vision.

Don’t know much about the tech other than what the video presentation talked about. How it works is something yet to be discovered.

To achieve such depth of field, and be capable of manipulating it after capture has challenged my entire perspective about how light works and what it actually is.

Did you watch the video at about half way through he manipulates a photo perspective… it’s incredible that a single camera source can gather do that.

Yes @PhiNotPi, watch the video. This is a single image, not a composite. It can have a shallow or deep depth of field depending on the desired effect. There is a tremendous amount of information to work with, much more so than standard imagery.

Well, that is a cool technology. (but it’s not magic!)

All I have to say is about damn time it was made available to the public.