Light Field camera - Is this a revolution?

[tara]

Dear All DSLR owner's,

There is new type of camera and few may be already know which focuses after taking the picture. Its so small camera and it looks fantastic for the price.

What do you guys have to say on this. Does this mean eventually all DSLR's will be gone? What is the future of SLR's if these tiny can refocus and get the same image quality like DSLR's

https://www.lytro.com/camera

Look for the picture gallary and click at different part of the image and see how can refocus.
https://www.lytro.com/living-pictures/282

Steve Lohr, The New York Times says :
"For a photographer, whether amateur or professional, the Lytro technology means that the headaches of focusing a shot go away."

Thanks for your opinon.

 

[tara]

I am little afraid, if all my lens will be waste if something like this makes revolution. May be the first version of this light field camera may not be as good to replace DSLR's.

But I was wondering in future there could be many manufactures who may jump into this market and after few revised version of this kind of camera the IQ and focus could be very close to DSLR's

What's your thought on this?

Thanks

 

[necator]

My take: No!

a) It's not new.
b) You are sacrifying resolution---a lot of resolution. (That's given by the principle of the Lytro.)

Even on their low-res test-samples you can see the noise.

Therefore: Interresting: Definitly
Valuable for specific solutions: Yes
Will make DSLR obsolete: Never

Of course, since I never had a Lytro on my own, I might be completely wrong...

 

[torger]

No, it is not a revolution, it is a toy. IQ-wise it is competing with mobile phone cameras.

 

[caruser]

No, it is not a revolution, it is a toy. IQ-wise it is competing with mobile phone cameras.

The Personal Computer (in its generic sense) started as a toy, too, and look what happened to the "real" computers.

The question is not whether, in its current, first incarnation it is a threat to the DSLRs or P'n'S cameras, the question is, extrapolating another 10 years of technological progress — is there any inherent reason that it will not be usable by Pros?

The digital camera and the phone camera initially got laughed at, too, because they started really low down, and look what happened to them...

 

[Forceflow]

No, it is not a revolution, it is a toy. IQ-wise it is competing with mobile phone cameras.

The Personal Computer (in its generic sense) started as a toy, too, and look what happened to the "real" computers.

The question is not whether, in its current, first incarnation it is a threat to the DSLRs or P'n'S cameras, the question is, extrapolating another 10 years of technological progress — is there any inherent reason that it will not be usable by Pros?

The digital camera and the phone camera initially got laughed at, too, because they started really low down, and look what happened to them...

Just by the way the technology is working the IQ of the Lytro will always be far lower than what you would get with a regular camera with the same sensor size. So no matter how much technology advances, the 'normal' technology will be ahead. Now it might be that at some point the sensors will get so good that the drop in IQ and resolution wont matter, but I do not think this will ever happen. Problem there is that at some point you will simply reach the limit of what is physically possible due to the way light behaves, and I don't think we are that far away from that border anymore.

 

[samthefish]

Not in its current form is it a revolution - you have to devote an entire array of pixels on the sensor to capture one "lightfield pixel" or whatever you would call it. For that effective resolution will always be a significant fraction of the native sensor - they're saying 11M "lightfield points" but the resultant image is only HD, probably something like a 1 megapixel image. The tradeoff they are making is one of resolution for directional info on the light. Now if they had a 110mp sensor and a resultant 10mp image perhaps it could be really stunning for more professional type photography. They're certainly targeting the "wanna post a cool pic to facebook" crowd with this, maybe if they make nice profits it will fund R&D into higher end cameras.

I think it's very cool what they've created though - a fixed 2.0 av camera with no need to focus. I actually read their CEOs research paper on lightfields and found much of it to be very readable.

One interesting thing is that with advances in CMOS technology it shouldn't be difficult to make very high resolution sensors - Canon and Nikon have (correctly in my mind) focused on have lower noise / higher dynamic range per pixel than cramming as many as they can on the sensor. Perhaps lightfield could make use of the ultra high resolution sensors possible now.

SamTheFish

 

[caruser]

No, it is not a revolution, it is a toy. IQ-wise it is competing with mobile phone cameras.

The Personal Computer (in its generic sense) started as a toy, too, and look what happened to the "real" computers.

The question is not whether, in its current, first incarnation it is a threat to the DSLRs or P'n'S cameras, the question is, extrapolating another 10 years of technological progress — is there any inherent reason that it will not be usable by Pros?

The digital camera and the phone camera initially got laughed at, too, because they started really low down, and look what happened to them...

Just by the way the technology is working the IQ of the Lytro will always be far lower than what you would get with a regular camera with the same sensor size. So no matter how much technology advances, the 'normal' technology will be ahead. Now it might be that at some point the sensors will get so good that the drop in IQ and resolution wont matter, but I do not think this will ever happen. Problem there is that at some point you will simply reach the limit of what is physically possible due to the way light behaves, and I don't think we are that far away from that border anymore.

Let me reformulate and use a different analogy. Nowadays most cameras don't have an optical viewfinder, because although the alternatives are not (yet) better, they are good enough for most people. Does this technology have the potential to become good enough and cheap enough to take over the point'n'shoot market (or what's left of it after the phones finish eating it for lunch)? Or will it be integrated into a phone too? ;-) What about EVIL or DSLR? Or will the new world be lightfield for the masses and DSLRs for the relative niche of ambitious and professional photographers?

 

[Mt Spokane Photography]

Right now, the new models that will come out will be low resolution. There will not be any advanced tools like light room or photoshop to deal with the images, but that may come later. Adobe had done quite a bit of research and just mayget on the bandwagon.

Great things have humble beginnings, so it depends on user acceptance. It it is popular with the mass market, then $$ magically appear to develop more sophisticated versions, or a $$$ company like sony may buy it up and develop it. If people start finding new uses for the technology, it will grow, otherwise it will disappear or reappear in a new form every few years.

There are lots of questions to be answered, like printing, web pages (apparently web viewing software is supplied).

 

[Loswr]

the resultant image is only HD, probably something like a 1 megapixel image.

At best, it's a ~3 MP image, and that's assuming an optimal compromise on microlens to sensor distance, as Ng described in his thesis. FWIW, based on several factors I suspect the camera uses a 1/2.3" 14 MP sensor (CCD or CMOS, both are common and inexpensive components). Since as you state, you can't get anywhere near the full resolution of the sensor because you're trading that for the ability to refocus, it makes sense that instead of providing real, comparative resolution, they decided to make up a new unit (the 'megaray').

 

[ippikiokami]

According to the research pictures I think you lose resolution by a factor of 4. Obviously with the way it works right now there is no way it can get the dynamic range of a standard sensor of the same size.

The technology is still a long ways off of being mass produced. You will have to put much larger sensors in even point and shoots to get the same type of resolution and dynamic range.

It's cool no doubt. But it's still got a LONGGGGGGG way to where it's actually more than a novelty

 

[branden]

Besides whatever technical reasons, the real reason this technology won't be used by professionals is because in a planned, well-composed photograph there's no reason to ever want the focus to change.

If you want an interactive exhibit, that's fine, but it is in addition to photography, not in replacement of it.

 

[Loswr]

Besides whatever technical reasons, the real reason this technology won't be used by professionals is because in a planned, well-composed photograph there's no reason to ever want the focus to change.

If you want an interactive exhibit, that's fine, but it is in addition to photography, not in replacement of it.

You might not want to change it, but you might want more of it...consider it for macro photography, where it would produce in a single exposure what would otherwise take a whole set of exposures followed by focus stacking.

 

[Renzokuken]

Yes. This is a revolution.

But it's at it's infant stage now.
It cannot replace a decent DSLR in the near future, but that does not mean it can't let say 10 to 20 years from now.

The ability to focus shots after the image has been taken ensures that nothing is essentially missed.
The great once in the life time moments, first baby steps, first kiss, first day at school can now be safely recorded.

Essentially, a camera is suppose to capture moments.
Some may complain that with such a camera, photographers would become even lazier, and just spam their shots, undermining creativity
But the way I see it, you still need to compose a shot, you still need to decide what should be included or excluded in the frame.
If you're truly a passionate photographer, the liberty to be lazy (focus after the shot) or not is entirely up to you and not what the camera decides.

If integrated to security/surveillance cameras, this technology can provide much more information to investigators and better help them crack a case.

There are much more uses for such a technology. But with it being at infant stage, this technology is still fragile. If the industrial giants feel threaten by it and decides to do something to compete against it, it might become unavailable commercially.

It may not yet be able to replace a full-fledge DSLR yet, but all in all, humanity should welcome such innovation and technological advancement.

Good Day
Renzokuken

 

[sanyasi]

I just put down my slide rule so that I could write this post. I ordered a Lytro yesterday. Given all I spend on camera equipment, paper, and ink in a year, $500 for what looks like an entirely new technology seemed very little.

Will Version 1 replace DSLRs? No way, for many of the reasons written already. But there seems to be something threatening to the pros and hobbyists about anything that permits the masses to create anything comparable to what the pros and hobbyists create. But what is going to keep this technology moving forward if it shows promise is not the pros and the hobbyists. Its the ad agencies, newspapers, magazine editors, and other consumers of photographs that will drive the adoption. If there is a better and cheaper way to get the job done, they will gravitate in that direction. We all say it comes down to vision, composition, and creativity. It most certainly does, but if a music reviewer can get a great photo, you can bet the newspaper won't send a photographer.

I suspect that once people look at the technology, they will find all sorts of way to use it for fine art and creative ways. Cameras are tools and my experience is that no matter what the tool, creative people find unintended ways to "misuse" it to obtain creative results.

The fundamental question comes down to this: If you had a fantastic idea and opportunity to take the photo of a lifetime, is the photo any less fantastic because it was easier than some other way or that others could do the same thing? What then is a great photo? Is it in the eyes of the maker or the viewer?

In my view, new technologies equals new opportunities. But no matter. People are still happily using film, Holgas, and view cameras and having a lot fun doing so. You will still be able to use your DSLR once the Lytro comes out. And you can still play your vinyl records. The Lytro is the perfect device--it keeps forums like this one going.

I will be curious to see whether Really Right Stuff makes a plate to mount the Lytro to one of its tripods. That is the advanced hobbyist in me speaking.

 

[kubelik]

I would love to have a Rebel with Lytro capabilities to take on vacation. it would be awesome to take photos that are well composed with the right background blur ... but when you get bored or curious, to then look and see what else is in the scene that you might have missed. especially if you're shooting urban photography or scenes of daily life, there could be wonderful moments discovered by having a camera with this sort of technology.

I hope it does well because I think it's a great idea that could be very interesting. I don't think it replaces having an initial artistic vision of how you want an image to appear, unless you plan on spray-and-pray and then tons of hours of scrolling through the DOF-slider to see if you've caught anything good.

 

[tara]

Thanks to everyone for their input. Its very much appreciated. Its so nice to see opinions.
Please do add your your thoughts.

 

[JonJT]

I think this could be a novel feature in a future DSLR body. I just want to be able to continue to use my current lenses. From the very small amount of reading that I've done, it seems that I just might be able to. Could anyone confirm this?

 

[archangelrichard]

too many people do not understand what this is and how it works. Part of that is the designer's fault - he sprew hyperbole

NO this is not a camera, per se; but the image capture portion of a computer software based imaging system.

What this does is take a very small (less than 1 MP) image on the microlens surface and each large sensor/lens divides that up into (the estimate I have heard most often is 11) microlenses each of which has a different micro-length

NO, there is no light field; that is just a hype way of explaining the number of different focuses

each focus is at F 2.0 Depth of Field so at very long lens images there are gaps in the focus at close range (areas that don't match any of the pixel/focus pairs)

The full image can only be viewed via their site (and how much do you want to bet that will go subscription) or, in theory on a Macintosh running OS X 10.6 or above and their software (and even then some things will require using their site

posting the image/grid requires linking from their site; they already have a facebook link, as all the software to display the image is proprietary on their site

since there is no single image, ever; only the grid / matrix of images you can't do any manipulation on them except at their site / what they offer

There is NO WAY to print a picture because there is no picture; just that grid / matrix

Because all the work is done by the microlenses, there is no way to have any real camera use this technology - you either get a microlens that is always at the same focus as it's neighbors for a 18 MP (for example) image or the varied microlens sheet in front of your sensor for a 1 MP grid / matrix

He also confuses the notion of having the information of the direction of light - understand that the purpose of a lens is to have only 1 direction from the front surface to the sensor thereby having the whole picture in focus (if such a thing were possible, all you would get is a colorful blur, chromatic aberration is what comes of different directions in the light / image from the lens to the sensor). It would be possible to make 3D out of this the same way hollywood does in the movies - by calculation based on where in the grid an item is in focus which would give you an idea of distance

This is vastly overhyped and some of that is outright false or misleading - what this is is a camera-toy for little high school (and pre-high school) to use for facebook; you know, the ones who post their status every two minutes; lots of music you can't stop, horrid assault-to-the-senses color scheme, and thousands of nearly identical pictures, thousands of "friends" most of whom they could not possibly have ever met

That is their market and this thing will compete with smartphones as the camera / gimmick of choice

NO this isn't a technology as much as an application of technology. NO you can't add it to any existing system of photography. NO, you really wouldn't want to (less than 1 MP?)

 

[AprilForever]

No, I don't think so... the computing power required for a "real" sized image from a "real" sized sensor would be ludicrous....

Besides, I looked at their website image gallery yesterda: Dismal IQ. Very not sharp. Colors look flat. Bokeh not too good. And the arrogance of their you-can-only-process-this-through-us policy throughly rankled me!

This is a gimmilck, to be toyed with by the unwashen masses....