Never take an out of focus shot again???

[Glen Roberts]

Anyone seen this before???
Apparently it's uses a different kind of sensor that reads the light in a different way to ordinary camera sensors. Thus enabling you to change the focus of a shot AFTER it was taken.

If you look at the picture gallery you can click an area on the shot to change the focus.

http://www.lytro.com/cameras

 

[Hamish Gill]

blimey ...
I wonder what the actual camera looks like?
that's quite impressive really...
I shall keep an eye on that i think

 

[Pete Askew]

I would have thought that they must have limited capabilities if the processing is being done in-camera. Look forward to seeing the reviews.

 

[Chris Dodkin]

This is going to put the cat amongst the pigeons if it works!

 

[Hamish Gill]

Yeah, I think your right!
I van see a lot of people being critical of it too saying it's not real photography etc
I think it's very interesting, but like most new tech, it will probably take a while to get good enough to compete with traditional technology ... ?

 

[Larry Bolch]

As I understand it, it requires a very large sensor, but just outputs a rather low resolution image. I am not sure who it is aimed at. A knowledgeable enthusiast will choose an appropriate lens and aperture for the effect wanted, and probably knows well how to focus a camera. While it might save a snapshooter from the plague of fuzzygraphs, choosing the point of focus would definitely require a computer and processing software—not something most snapshooters are into. A touch-screen monitor on the camera could be a reasonable compromise.

Considering that the shipping date is somewhere in the vague future, I expect that the current hype is aimed more at venture capital than at potential customers. Vaporware is the term. It smells much like the announcement of the Foveon sensor that I was assigned to cover for a Brit hi-tech magazine. According to Foveon's hype-machine, every camera maker in the world was beating upon their door, and it was just a matter of days before we all threw our dreadful Beyer mosaic CCD based cameras in the trash and blissfully shot with Foveons forever after. That was not exactly the way it worked out.

Real photography is creating an image using light. The term could not be more all encompassing. I recall a decade and a half back just as image capture was becoming electronic, crusty old dolts claiming it was not photography. Photography required film. Wrong then, historically wrong—and wrong now. If the end product is an image made with light—it IS a photograph.

 

[Larry Bolch]

A second though just occurred to me. With the power of camera processors and decent sized buffers I can think of no reason why focus bracketing could not be implemented. The camera can calculate the depth of field at a given aperture, and shoot a series that would have somewhat overlapping sharp images. At the most basic, simply select the one you like best. But now Photoshop has stacking, something long used in astronomic photography. It can combine the images into a single image with depth of field from the nearest point to infinity. For the P&S set, this could be done in-camera as panoramic stitching is done now.

I have tried this with an object close-up and a series of shots with focusing distance varying. It chose the images with the highest edge contrast and seamlessly combined them. It will also let you shoot a series on say, a busy street. As long as the people and cars are in different places, they will disappear, leaving only the street and buildings that appear in the majority of the images.

Not an elegant approach with leading edge technology, but entirely practical and could be implemented now.

 

[Pete Askew]

There is already one of these type of camera on the market but it costs about $30K. This is supposed to be in the consumer price level but will presumably be a lower resolution device. Stiil it is interesting technology and has some practical applications (sort of like con-focal microscopy) including imaging partially occluded subjects. Can't get too excited about is a creative tool and there are significant technical problems still to resolve. Let's see where it goes.