LYTRO Light field camera

Sounds like marketing crap.

"Click anywhere to focus" Not much point when your focusing on a soft as hell image.
 
fixedimage said:
"Click anywhere to focus" Not much point when your focusing on a soft as hell image.
Click to expand...

I though that was the point of it - you can chose to focus anywhere after the event. If it works of not and proves to be more then a gimmick remains to be seen. There's some restrictions to its use at the moment so be interesting how it turns out
 
I only got as far as the first page, my anti virus blocked the rest.
I can't see much use for me, I know where I want the focus to be, and as said the whole image is soft, even a cheap lens is better than that.
 
I'm not sure the design is right. Even if they can make it that small everyone is comfortable with the ergonmics of a "standard" camera so why not package it as such?

"Mega-lightsrays" is surely marketing jargon in an effort to make it sound like megapixels?

Time will tell
 
So it takes an image at f22 then "powerful" software adds a Gaussian blur to all but the selected area's?? I think sony are releasing something almost identical in a normal looking camera.. Re-inventing the wheel but using marketing and software to have us believe it's some sort of newfangled technology..
 
fixedimage said:
Yeah but when it focused it was exactly sharp.
Click to expand...
No it isn't. check out the surfer, click the focus point in the water nothings even close to sharp. And thats on a 5x5 image, heaven help a 20x16.
It's not focusing on distant objects well.
 
Last edited:
Not much confidence in this technology then. It is still in its infancy so I guess we will just have to wait and see.

Perhaps we will see in some future version of lightroom or PS an option to pick a focus point from our RAW images:lol:

Personally I see this as a gimmick which many consumers would by into, so I can see this creeping into some compacts / point and shoots in the future.
 
Read the dissertation if you're tech minded. The math is a bit too much for me but the basic principle is well explained.

And no it's not taking a normal picture with a small aperture and blurring it. How would it know the distance of each pixel from the camera to even do that?

Basically it uses several sensor pixels + slanted micro lenses to record multiple directional color values per image pixel, thus giving the ability to simulate a focusing lens in software. The camera itself has a 8x zoom at fixed f/2 but of course no focusing.

It's not marketing crap. It's the first generation of a mass market application of light field cameras. You can google for the research versions which fill a room.

The RAW files seem to be huge - 20+MB each and the resulting image is probably closer to 1Mpx.

A gimmick today but give it 10 years.
 
Where would we be without face recognition? .... Pro togs should be shaking in their boots! .... Marketing crap :thumbsdown:
 
swanseamale47 said:
No it isn't. check out the surfer, click the focus point in the water nothings even close to sharp. And thats on a 5x5 image, heaven help a 20x16.
It's not focusing on distant objects well.
Click to expand...

Yeah, sorry my post should have read "WASN'T exactly sharp"
 
vaizki said:
A gimmick today but give it 10 years.
Click to expand...

I agree. Against modern point n shoot cameras, it seems to be an over engineered fix to a problem that doesn't really exist. With a touch screen camera, it is much easier for the user to select what they want in focus before taking the photo. Having to use special software to view the photos also makes it seem gimmicky in its current state.

But I could easily image such a sensor being used for video in the future. A 3D display teamed with a small camera to track your eye movements would allow for great immersion as whatever you are looking at would be in focus. With current 3D films your only reward is a splitting headache if you look at something other than what the director has chosen to focus on.
 
Last edited:
vaizki said:
Read the dissertation if you're tech minded. The math is a bit too much for me but the basic principle is well explained.

And no it's not taking a normal picture with a small aperture and blurring it. How would it know the distance of each pixel from the camera to even do that?

Basically it uses several sensor pixels + slanted micro lenses to record multiple directional color values per image pixel, thus giving the ability to simulate a focusing lens in software. The camera itself has a 8x zoom at fixed f/2 but of course no focusing.

It's not marketing crap. It's the first generation of a mass market application of light field cameras. You can google for the research versions which fill a room.

The RAW files seem to be huge - 20+MB each and the resulting image is probably closer to 1Mpx.

A gimmick today but give it 10 years.
Click to expand...

By which time it will be ten year old crap ;)

They've trawled all the major photo brands and got zip. It's just a marketing attempt to get some gullible manufacturer to take it on.

Google Nimslo 3D from the 80s for the last photo-con that nobody wanted. That actually got to production before reality dawned.
 
HoppyUK said:
By which time it will be ten year old crap ;)

They've trawled all the major photo brands and got zip. It's just a marketing attempt to get some gullible manufacturer to take it on.

Google Nimslo 3D from the 80s for the last photo-con that nobody wanted. That actually got to production before reality dawned.
Click to expand...

20 years laters and we have the Nintendo 3DS, HTC 3D Evo, Fuji FinePix 3D...
 
Last edited:
HoppyUK said:
By which time it will be ten year old crap ;)
Click to expand...

And I seem to remember something similar being said about this new fangled thing using digital sensors to capture images that would never replace film. It never caught on, did it? :lol:
 
Rod.F said:
20 years laters and we have the Nintendo 3DS, HTC 3D Evo, Fuji FinePix 3D...
Click to expand...

3d is a fad. As it was in the 80's when it suddenly died a death when people got bored with it! I've seen a few 3d films of late and to be honest, after 5 minutes you completely forget its in 3d and all you're left with is a film which looks underexposed for 2 hours as you're forced to leave the glasses on!

The only area where 3d will take off IMO is gaming.
 
Alastair said:
And I seem to remember something similar being said about this new fangled thing using digital sensors to capture images that would never replace film. It never caught on, did it? :lol:
Click to expand...

But digital had so many advantages, this was obvious from the outset of digital sensors. This system doesnt really does it? Anyone worth their salt can easily focus on what they want to focus!

Have people forgot we have AF?! :lol: If you miss focus, take another shot! Really, I cant see how this system can be an advantage. The focus is all part of the composition.
 
Last edited:
odd jim said:
But digital had so many advantages, this was obvious from the outset of digital sensors. This system doesnt really does it? Anyone worth their salt can easily focus on what they want to focus!
Click to expand...

Was it? I remember trying to sell the original Mavicas when they came out about 15 years ago (the ones that took a 3.25" disk as memory media and just about made VGA, battery lasted for an hour or two), there was no real conceivable advantage at that time (they didn't sell well). It's not that long ago that digital became dominant in the professional market (I would guess at not much more than 5 years ago), there was (and still from some people) strong resistance to digital on the basis it offers nothing but convenience.

But I think the problem is that thinking about the Lytro with a conventional camera mindset is like thinking about the new Nikon mirrorless from the point-of-view of the typical DSLR user. It's not about persuading existing modes of thought, it's about trying to change the game.

Lytro will either stand or fall, but it won't be on the basis of comparison to conventional imaging technologies. If it takes off, it will be because it changes the way the people that buy into it use or think about the images it produces. It won't change the way people use conventional cameras. And if it fails, then in ten years time it will be back as the new Holga..
 
Rod.F said:
20 years laters and we have the Nintendo 3DS, HTC 3D Evo, Fuji FinePix 3D...
Click to expand...

Alastair said:
And I seem to remember something similar being said about this new fangled thing using digital sensors to capture images that would never replace film. It never caught on, did it? :lol:
Click to expand...

As Jim says above really.

3D has been around since the dawn of photography and comes and goes at regular intervals. Nobody has overcome the problem of viewing because you need two of everything to make it possible - for both shooting and viewing. For as long as we have two eyes.

It's great for gaming, but nothing else. 3D cinema audiences are dropping and I don't know anyone who has bought a 3D TV or camera, though I guess they must exist. Basically, nobody wants it and our brains are very good at recreating a 3D image in our minds from a 2D scene.

I don't think anybody ever said digital wouldn't catch on, if the quality could be delivered at the right price. And when it did, it immediately swept film away. Obvious. Edit: Sony's Mavica, from the early 80s, was not the breakthrough digital camera - it had the wrong technology (basically video tape on a disc) and could only manage 0.3mp resolution which was never going to get very far.

The problem with Lytro is that it does something nobody wants, the ability to refocus an image after it's been taken, when it's not hard to do that at the time of shooting. And as far as the mainstream consumer market is concerned, they use compacts and phone-cams where everything is sharp front to back anyway.

That Lytro guy talks the kind marketing mega-[PLEASE DON'T TRY TO BYPASS THE SWEAR FILTER] you can drive a coach and horses through sideways. It's full of buzz words, meaningless hyperbole and ridiculous assumptions and contradictions. The only thing he doesn't say is that if he can con(vince) a major manufacturer into taking it on, he'll make a fortune. That's the only reality.

The other thing is image quality, which is feeble because Lytro effectively shoots a whole range of images at different focusing distances simultaneously so you can choose the one you want later. The downside is that sensor resolution is dramatically reduced, effectively divided by however many focusing zones there are.

It also occurs to me that if this is a desirable feature, then something like the new Nikon 1 could shoot an extremely rapid sequence of high-res frames (it already does 60fps) at a range of focusing distances and do the same thing - using existing technology and with far higher quality. The problem is, nobody wants it.
 
Last edited:
HoppyUK said:
It also occurs to me that if this is a desirable feature, then something like the new Nikon 1 could shoot an extremely rapid sequence of high-res frames (it already does 60fps) at a range of focusing distances and do the same thing - using existing technology and with far higher quality. The problem is, nobody wants it.
Click to expand...

That sounds like a useful feature for any subject where DoF is a problem - like macro.

It's not a new idea though.
From this review of the Canon G9 in 2007 - "The G9 also provides Auto Focus Bracketing (AFB). Enable AFB and with a single push of the shutter button, the camera captures three exposures in rapid succession marginally shifting the focus for each (one just slightly in front of the optimum focus point, one at the optimum focus distance, and one just slightly behind the optimum focus point), virtually guaranteeing at least one correctly focused image even in rapidly unfolding action situations"
 
DuncanDisorderly said:
That sounds like a useful feature for any subject where DoF is a problem - like macro.

It's not a new idea though.
From this review of the Canon G9 in 2007 - "The G9 also provides Auto Focus Bracketing (AFB). Enable AFB and with a single push of the shutter button, the camera captures three exposures in rapid succession marginally shifting the focus for each (one just slightly in front of the optimum focus point, one at the optimum focus distance, and one just slightly behind the optimum focus point), virtually guaranteeing at least one correctly focused image even in rapidly unfolding action situations"
Click to expand...

Yes, another variation on the same theme is focus stacking technique commonly used for macro where multiple images are taken at different focusing distances and combined in post processing to greatly extend the depth of field.
 
HoppyUK said:
As Jim says above really.

3D has been around since the dawn of photography and comes and goes at regular intervals. Nobody has overcome the problem of viewing because you need two of everything to make it possible - for both shooting and viewing. For as long as we have two eyes.
Click to expand...

But Lytro comes the closest to providing 3d without two of everything.

I could see it catching on, but it needs some development on the resolution and the ability to record a moving image. A drop in the cost would also help, but if the resolution can be enhanced at the same price point that shouldn't be an issue. I can already see possibilities (although not $399 worth at the moment).

I'm racking my memory to find something earlier, but is this the first example of an essentially unprintable still camera technology? i.e. the intrinsic nature of the media can not be captured on a piece of paper, you can only print a snapshot of piece of the total data captured.
 
3d tv will take off when glasses aren't needed and have wide viewing angles ie same as normal tv. Until then it's a useless pain in the arse.

The lytro idea sounds interesting and I reckon it will have the same culty following that lomography attracts.

Would be ideal for wildlife. Would only have to vaguely wave camera at BIF and have hope in hell of getting it in focus later :)
 
Alastair said:
But Lytro comes the closest to providing 3d without two of everything.

I could see it catching on, but it needs some development on the resolution and the ability to record a moving image. A drop in the cost would also help, but if the resolution can be enhanced at the same price point that shouldn't be an issue. I can already see possibilities (although not $399 worth at the moment).

I'm racking my memory to find something earlier, but is this the first example of an essentially unprintable still camera technology? i.e. the intrinsic nature of the media can not be captured on a piece of paper, you can only print a snapshot of piece of the total data captured.
Click to expand...

Lytro's claims about 3D potential are part of the nonsense. What I think they must be referring to is some kind of simulated 3D where two images are produced at different focusing distances, one viewed by each eye to synthesise some kind of jump-out effect. It's not 3D, but they are very vague about it, as they are about everything else.

I don't understand your other point.
 
Has anyone actually read and understood the thesis? http://www.lytro.com/renng-thesis.pdf

The optics and physics seem to be good at first glance

They are using an array of micro-lenses, and doing maths on all the images those micro-lenses produce - based on reverse ray tracing. They are looking at how objects at different distances away form different images dependant on what part of the optical plane the image is formed from

It was a fascinating read, and I reckon the idea expounded in the thesis in itself has a lot of engineering applications i.e. anything that needs accurate 3d tracking without refocusing
 
Last edited:
Let's be honest, if it was any good one of the main players would have bought the patent off them by now......
 
Last edited:
Richard King said:
Has anyone actually read and understood the thesis? http://www.lytro.com/renng-thesis.pdf

The optics and physics seem to be good at first glance

They are using an array of micro-lenses, and doing maths on all the images those micro-lenses produce - based on reverse ray tracing. They are looking at how objects at different distances away form different images dependant on what part of the optical plane the image is formed from

It was a fascinating read, and I reckon the idea expounded in the thesis in itself has a lot of engineering applications i.e. anything that needs accurate 3d tracking without refocusing
Click to expand...

Yes. But that's not how the technology is being pitched to the market. A cynic might say that there's not much money to be made out of selling a handful of cameras to engineers and scientists...

Edit: I'd not seen that link before Richard, so thanks for that. But you don't have to go beyond section 1.1 The Focus Problem in Photography to know that the basic assumptions on which the entire thesis is based, are fundamentally flawed. It's a wish-list of untruths and misunderstandings masquerading as science.
 
Last edited:
Richard King said:
It was a fascinating read, and I reckon the idea expounded in the thesis in itself has a lot of engineering applications i.e. anything that needs accurate 3d tracking without refocusing
Click to expand...

It is a very cool technology; but it's not new....
Raytrix have had a commercially available product for a couple of years.
Check their home page; their product does exactly what you stated.
The difference is that Raytrix have targeted engineering applications, whereas Lytro are going after the consumer market.
 
I never claimed who developed it, and to be fair in the dissertation neither do Lytro

I never said I thought it was a fab idea for photography either

As a person who just loves maths and Optics. It was a very good read, and at face value, it's an idea that has a lot of potential, even for conventional phootgraphy - i.e. how about for actually really good 3d focus tracking for 2 D cameras etc. Or for use as a trigger to detect movement towards you of a subject etc.
 
srichards said:
3d tv will take off when glasses aren't needed and have wide viewing angles ie same as normal tv. Until then it's a useless pain in the arse.

The lytro idea sounds interesting and I reckon it will have the same culty following that lomography attracts.

Would be ideal for wildlife. Would only have to vaguely wave camera at BIF and have hope in hell of getting it in focus later :)
Click to expand...

You can buy these TVs already where they produce the full 3d image on their outer panel, Sony, Panasonic and now Samsung make them. The Sony has been out a while and needless to say, sales arnt booming!
 
Last edited:
odd jim said:
You can buy these TVs already where they produce the full 3d image on their outer panel, Sony, Panasonic and now Samsung make them. The Sony has been out a while and needless to say, sales arnt booming!
Click to expand...

Possibly because you have to sit motionless at a specific distance from the screen, and they give you a headache!

Good for gamers, but I think the new systems for them now carry a health warning not to watch them for long :eek:
 
Did video ever kill the radio star :shrug: There will probably always be a use for holding a lump of plastic and glass in your mitts to capture an image, but at some point in the future people will be using smaller and more sophisticated pieces of kit. Hey! even a panoramic add-on for your phone :lol:
http://laughingsquid.com/dot-a-360-degree-camera-attachment-for-iphone/
 
I like the idea of it but it's the first commercial version. the first digital cameras were only around 1mp and were not cheap. look how things of changed in 10 years.
 
haha it's the iPad announcement all over again :lol:

remember when the iPad came out and everyone was like "that's stupid I have an iPhone and that's actually pocketable" and look now, pretty much everyone who made that statement now owns an iPad and don't know where they'd be without it.

Lytro are just doing it wrong, instead of making a camera for consumers they should target commercial aspects, I don't know what industries it may appeal to apart from possibly the engineering/manufacturing and even healthcare side of things.

It's great technology it just needs to find it's purpose.
 
Nick T said:
:wave:
You do now :)

Mine's awesome - 3D films look fantastic through it and playing Arkham City in 3D is just stunningly good.

:)
Click to expand...

:thumbs:
 
I get so many focus misses that something like this would be awesome! I can't be the only one... Can I? Oh dear... I'll get my coat.
 
Couldn't manage the full dissertaion but just scanned it - it's an interesting read.

The technology looks interesting for definite. I don't know how it will be best applied and who by, but it will have its uses I'm sure. It'll probably get better and better over the next few years and when we're all using them we'll wonder what the issue was....;)
 
You must log in or register to reply here.
Back
Top