Lytro plenoptic camera used for fashion shoot

These images, as do other samples I've seen, seem to have 2 or three focus zones at maximum. The most I've seen in a sample is 4 focus zones.

Nice thig

what is the usage?

Well, it just OK for fun. Very interesting!

Dynamic tomography - just like airport scanners.

@GordonAtWork. The real thing - if it exists - can do much more than this. It will have an infinite number of focus planes and it will be able to have an arbitrary DOF. These demos are just lame. If they are fake or not I dont know. But I am quite sure there are no real camera. Its probably a lab prototype based on a FF Sony camera or Canon 5D connected to a laptop computer.

For me, all of the hoopla now is just that--PR hoopla and puffery. I'll believe that the product really does what its company says that it does when they make a sample of the shipping product available to dpreview.

I'd like to see some images with a lot of focal planes rather than what seems to be just two.

FAKE! FAKE ! FAKE!
These are note real pictures!
Look at the picture of the girl with the city behind her!
You click on the girl, the city becomes blur. OK. But you click on the handrail, at the bottom right, which is obviously closer to the camera than the girl. The handrail becomes clear, but so does the city behind and the girl becomes blur. This make no sense! So it is obvious that the flash player animation presentend on this site are not real images!

Lots of problems with this, despite the wizzy demonstration. Looking between the lines Lytro suffers from:
- Huge data files but very low resolution
- The need to post-process to produce a useable file or print
- Even the in-focus bits are blurry!
- Not a patch on even basic P&S cameras

Most P&S users take most of their pictures of friends and family. Their major fruatration has, in the past, been missing focus on people's faces. Despite being derided as a "gimmic" in these forums, I know lots of people who think face recognition in compact cameras has massively increased their number of usable and printable shots.

Fundamentally the biggest market this camera has, reducing the number of out of focus shots of people, has been solved with face recognition. Sure you get a faux artistic background bokeh blur, but photographers looking for that "artistic" shot won't be prepared to put up with the camera's other sever limitations.

Hi, I'm the "snake oil" commenter. I'm not freaked out about this, I welcome it. But when all you see is a few very low res images, no images of the camera, and no details on how the pix were captured it makes me suspicious. How long does it take to capture an image? Will the camera use "standard" shutter speeds, or is something else required? What will it cost? Can you change lenses? I love the concept, but right now we all know next to nothing about the camera and its capabilities. Still vaporware, and I've been around long enough to not get excited at vaporware, no matter what the promises.

If they can get the resolution of the final images up to 6MP it should be good. I am curious as to how the low light & burst performance on this will be. Raytrix cameras use Nikon's F-mount which is also awesome.

I don't get why people are so freaked out about change, this seems to have all the upsides of DSLRs, plus the added bonus of not having to worry about focusing, with none of the downsides. People crying about how it will "kill the art of photography" are dumb, you still have to compose & process the pictures which is the most important part. I am certain people were this freaked out when DSLRs came out... "$5000 for 2MP? No thanks I will keep processing photos in the $10,000 darkroom shed I built on the side of my house"...

Purists ruin every hobby I am in to. You guys can continue with your manual focus lenses & film... that's fine... but don't knock everyone else for transitioning into the 21st century.

great idea, can be a great toy

but i still like my normal SLR

At 1:26 in the video, I think you have an unblurred (though tiny) sight of the camera. It seems to use a lens with rather large diameter.

494x494 (or 0.24 MP) would be a typical resolution for conventional Plenoptics refocus. Technology and market leader Raytrix from Germany (read more about the market here: optics.org/indepth/2/6/3 ) hold all the patents to push this higher up (currently up to 3 MP with their R11 camera), cf. raytrix.de/tl_files/downloads/Raytrix_Intro_LightField_100124_web.pdf page 27 and compare with the red line for a standard plenoptic camera which goes thru approx. 0.2MP.

So, the Lytro is easy to do and what is first is that somebody now thought it would actually make sense. But only America can turn this into a hype ;)

whaaaat? a camera that stores the photos on a flash memory?? people will never buy this hehe

A trick of 60 FPS Casio, with focus bracketing...?

lame.
gimme a proper lens with manual control any day.
i will never be interested in this.

I guess it´s similar technology that will come with Pentax Q for DOF effect

To all the skeptics, see you in 5 years from now. If it was a public company I would like to buy a pice of it.

Given a little time and an ability to produce full screen 1:1 size images I can see a phenominal aplication environment in security and forensic fields of photography and video closed circuit recording. It was fun playing with what is there already though.

Eventually having depth information for each pixel, in addition to color, will really enhance selections and manipulations in Photoshop. It should make selecting complex foregrounds or backgrounds that much quicker and cleaner.

This technology has been around for decades in some forms. In 1984 I worked to computerize a Dynamic Tomography process that could take a few simple xrays of an object from certain angles, then with special processing, put the slides together on a viewer and dial down through the object with very detailed resolution in all 3 dimensions. That program led to the scanners used in airports today. Getting all that muscle into a small camera is a trick....

As applied mathematics, this is awesome. As a practical way of producing digital images, it does not seem marketable yet. Interesting to see how long it will take before we can, and will want to, buy it.

What's a plenoptic camera?

The plenoptic approach has an advantage over focus bracketing in that it will not be affected by moving subjects in the way that multi-exposures are.

However, I don't think there is much of a market for it. Most serious photographers can focus anyway, and for those shooting in situations where it might be useful want better resolution. P&S owners just want everything in focus. Sure, there will be some who will be seduced by the marketing and buy one, play with it for 15 minutes and decide it is too much trouble and go back to their old camera. But not only that, cyberspace will soon be flooded with twitters expressing their disappointment - and the bottom falls out of the market.

snake oil is the best description yet!!!

Eric Cheng may indeed have developed some sort of plenoptic device. However, the equivalent effect could be achieved by means of:

1) A single camera with focus bracketing.
2) Multiple cameras or a single device with three or four lenses and sensors set to different depths.
3) A a deep focus camera with gaussian blur applied to the pictures.

#3 would be cheap and allow the greatest customization.

Any one of the options could be embedded in a device, and sold under the label of "plenoptic camera," without the buyer knowing any difference from the theoretical "real deal" described in Cheng's dissertation.

3D photography or video has more "wow" impact, but hasn't yet excited masses to buy 3D.

Eric Cheng is smart. If he applies his mind to create a 4 x 1/2.3" sensor camera that, with four lenses, allows a "spider-eye" cell phone to take pictures on par with a DSLR, he could make a fortune.

All of the Lytro pictures i've seen only seem to have 2 or 3 levels that it focuses at. I don't see how this gives any kind of advantage over a traditional camera with focus bracketing.

Good snooping DPR, I hope to find out more. 0.25 MP isn't that bad either ;)

Spam? Really?

OK, here's my take on the Lytro:

Still no images of the camera: why?
Still no hi-res pix: why?

I remain very skeptical. This seems to fall into the old category that if something is too good to be true, it isn't. Until I see a working camera and some full resolution pictures, this is snake oil.

This is brilliant! Provided the final picture quality satisfies at least actual prosumer DSLRs, I can see myself wanting one; instead of! Just need to hold it first and see the output IQ after manipulation. Hope it will be affordable, too...

Well, you can build your own plenotic camera, here it's shown how:

http://cameramaker.se/plenoptic.htm using an existing camera
http://cameramaker.se/microlenses.htm

Nowhere on the lytro site there is an image of a lytro camera. This means that there is no camera yet. The images are probably taken with some kind of prototype that looks like some optical and electronic experiment. I assume all software still is executed in a laptop. The conversion in software on this laptop may take minutes for all we know. There is not even a mock up concept camera built. All this tells that they are in a very early state regarding coming to market.

That fact does not say it is uninteresting. There are today actually cameras that do panorama stitching in software in the camera in real time. So - tomorrow will show.

Well ... the whole idea is great!
It will be interesting to view some comments on an article where photographers were discussing the first digital consumer camera.

I wonder how do you copyright your image? Hmm .... can you even put a watermark? Embed copyright matadata? You know, just some very basic practical things you do before going 'live' online, if this is for social media.

Am I missing something? It seems to me all you would need to achieve this is a binocular image (with a large DOF) and software. Depth information is encoded in the binocular displacement.