Anyone have any thoughts on the Lytro camera? The summary is basically that it records an image that can be focused after the fact.

Useful, or just a gimmick?

It seems to me that while this camera might be … underpowered … the idea is great. If you could combine the technology with higher-quality image recording and not have to worry about focus when taking a picture, that would provide more opportunities for a photographer to capture the image he wanted, especially in quick-response photography.

Useful, definitely not a gimmick, and something that they should really consider licensing out to manufacturers who already make good prosumer cameras instead of trying to wing it on their own. It seems like a waste of resources to have them working on their own firmware, physical design, marketing, etc. when all people really want is the technology in the cameras they already own (modified to use the new tech, of course.) I would pay a couple hundred extra for a camera with this built-in, and I reckon pros would pay many times that.

The Lytro is definitely a gimmick product. It shoots only tiny little images that are useful only with their player stack and contain only a handful of focus points, which can't be chosen.

This means you won't be able to do much with the images other than post them to Facebook. You may not get any one image to focus at all on the part you consider the most important to be in focus.

Apart from having a super-powerful incarnation of this technology able to create images with the fidelity of the top SLRs with many more focus points, allowing the shooter to pick the primary, I don't really see the point for general photography.

How many times are you ever going to refocus an image? It adds a lot of hassle, and in its current incarnation, requires an additional post-processing step that most people just aren't interested in.

As far as licensing goes, there are already other companies with higher-end solutions to this problem. The Lytro is likely borne out of the inability to convince any existing camera manufacturers to jump on the technology yet.

When 3D isn't flying, I'm not sure this has much of a future in the consumer space - not for a few years yet anyway. Too much hassle, little to no benefit, poor quality images.

Well clearly I do with the technology as according to Mark in the other thread I can't focus an image.

Until I get my hands on one I will reserve judgement, but I suspect it just takes a series of pictures and blends them together in a very limited fashion. I look forward to having a play myself as I find that is much better way of making an informed opinion.

Cheers

Cris

Heh. Two responses to Bitt, and two different opinions. I agree with this one. It's a gimmicky product.

Lightfield technology, however, is incredibly fascinating, and I could make great use of it, if it existed in pro-sumer camera models.

If the question is how often am I going to refocus my own pictures, well, hopefully never. But reality says I can sometimes use a post-photo re-focusing. A lot of what I've been shooting lately is live-action theatre. You only get once chance at an actor's expression, because their performance is never identical. There have been plenty of times where I've gotten what I think is a good shot (composition, lighting, and by review on the LCD, focus), only to get home and discover that the focus was on their shoulder, instead of on their face, and the depth of field means their face is slightly out of focus. It would be nice to be able to fix small errors like that, instead of throwing out an otherwise good shot.

But if you consider how often am I going to let someone else refocus my photos? Never. But I can imagine such technology being useful for things like virtual museum tours, or gigapixel panoramas, etc.

I saw a demo of this type of thing several months ago, a video from an electronics expo convention, and in the demo, the camera doesn't simply take a series of images from what I understand. It's got a different kind of approach to the lens and the sensor which means that everything is captured at the same instant, but with a range of available focal depths. I think the closest analogy was like that of an insect's compound eye, or something like that.

What I'm wondering is why, in the image playbacks, I only get the choice of focusing one plane of the image. Why not give me the option of all planes focused (and make that the default for goodness' sake).

No. It uses a different type of sensor to record the light. Did you poke around the site? They "explain" the science, and have some example photos up. There are some photos that make no sense (to me) to be refocusable, but there are also photos where changing the focus gives a very different mood to the photo, despite the subject matter remaining constant. As a photographer, I think the ability to explore those moods after a photo has been taken, is a very compelling feature, allowing you to be very precise about the image you convey.

So... the Lytro, gimmicky. Lightfields, not gimmicky.

Exactly. The Lytro's player allows anyone to refocus what appear to be simply bland/bad photos from their demo showcase I first saw last month (or was it September?)

The technology uses prismatic micro-lenses to capture multiple light fields, it doesn't take multiple photos. Its resulting file isn't exactly a photo either - it needs to be processed into one. And the Lytro implementation only captures sufficient light fields to produce a small number of focus possibilities. I believe it's 8 planes.

This isn't infinite focus by any stretch and I believe we're many years away from the technology being that mature, let alone being able to control depth of field which is a lot more fun than simply altering the focal plane. In the Lytro images I looked at, there wasn't much point to changing the focal plane - the images could have been captured with a traditional process producing better looking photographs. The Lytro only added the Facebook gimmick - which will be fun for about 5 minutes.

So basically the camera take 8 photos at the same time on fixed focal planes and combines them into a meaningful image in software that the user can then tweak between the planes but not actually fine tune the focus in any given plane.

So by reckoning that means all they have done is to figure out how to split the sensor up into sections with light focus differently across it's face to basically build 8 pictures taken at the same time. Like I said, it's just taking multiple pictures at different focal lengths from the exact viewpoint at the same time. With some very clever software working it all out.

I'm not impressed with the website of the examples they give. None of the images really work very well and you clearly can't fine tune focus and everything seems to be slightly out and low res. Still, I look forward to having a play.

Cheers

Cris

I looked thru the doctroral thesis of Lytro's CEO.

Lots of math. (way beyond me) but the basics are that you take a many megapixel sensor, the same as used in many cameras, and install an array of lenslets just in front of the sensor. This has the effect of creating an array of many small images focused on the sensor.

Each sub image covers a dozen or so pixels. All are slightly different from each other. The full result can then be post processed to generate an image. Adjusting various parameters in the math allows reconstruction of the image at various focal distances.

One issue in building the camera is, there is a relationship between the image sharpness that can be achieved vs. the ability to resolve the focal depth, is tied to the distance between the lens array and the sensor. The thesis has examples.

Another interesting bit is, since standard sensors are being used, there is no reason that video can't be captured. I'm guessing that the amount of post processing needed for video is extreme.

More megapixels would seem to be the answer to things like small images and sharper images with the ability to selectively focus thru greater depth.

If photography, from the very beginning, had been able to resolve everything in focus, a lot of what we call the photographic art, might not exist.

Piss off, Chris. Please do not attribute things to me which I did not actually say or post.

Cheers

No. You're not stirring the pot at all....

They are already using an 11 megapixel sensor to get (roughly) 500x500 images. Even if they magic'd up a 44 megapixel sensor you are still only looking at 1000x1000 images, which sure sounds at the gimmick end of the scale to me.

I've missed you Rob.

Sorry, how dare I doubt The Lord...



If you are going to insult me, at least have the decency to spell my name right.

Cheers

Cris

Here here! I miss the voice of reason.

Cheers

Cris

My two cents: The Lytro isn't a replacement for a traditional camera, particularly if you want to make big prints. However, it's a harbinger of some pretty interesting things to come, including non-traditional lenses that might bolt onto traditional cameras.

For what it's worth, I'm not getting one. If it were cheaper (say $100), then I'd get one for my daughter in a snap, since it would be a cool toy to play with.

At 500x500 resolution, I'd change that to "any prints"

500x500, IMO, isn't even high enough resolution for making images to post on the web.

Yea, I would agree. But it leaves me thinking what is the point at all ???

If it offered a low resolution image that I could fine focus and use it to extend the composition options in a shot then I would be very interested. At the moment I am struggling to think of many situations where I would actually want to focus on different planes. Maybe a bride walking down the aisle, perhaps.

Still I'll jump at the chance to have a play when I get the chance.

Cheers

Cris

This model looks to me like a technology demonstrator. I assume they'll sell a bunch, which will possibly appease their investors. Then in a few years time, after it's gone through a couple of generations of development, one of the existing camera manufacturers will license it (no doubt the technology is already suitably patented up) and incorporate it into a P+S.

I think Andy is spot on here. The startup environment right now is one that forces a little more then just a good idea and technology demo. Investors are looking for more solid things to invest in, resulting in startups having to turn into full on businesses to prove themselves. The technology also needs to evolve, and one way to help do so is to build it out and learn lessons from that process. Much like many things in life, this is one of those iterative situations. Digital cameras date back to the 70s, but took decades to mature to the point they became common consumer items. The tech behind Lytro and similar tech is also probably a decade out from common use, and these are the first baby steps.

Well in that case they are doomed to fail. I don't see their market.

People still want to print pictures, maybe less than they used to but they still do. Who wants a camera that you can hardly print from these days ???

Sure if it's taking interesting and different pictures I can seeing that being a drawn enough just to put them on an iPhone for example. But it doesn't (without me having a play with one) see to deliver on the creative element.

Cheers

Cris

The market I see isn't in the "put refocusable images on the web page". Thats just a shiny way to show of the tech. The future market is "Buy a digital camera that you never have to worry about focus again at the time of the shot". Point, shoot, then adjust later if necessary. This also has the advantage of letting the camera perform autofocusing in the background after the moment has been captured, instead of when the user is pressing the trigger button. The quicker a camera can go from off to picture taken, the better the chance the moment was captured as intended.

Thats my long term vision for it anyhow. Business plans can change much quicker then the technology behind this, so it's much easier for them to try a ton of different strategies to get this to catch on.

I'll point out that people have been decrying megapixel expansion for years. Here's something that could conceivably make those overlarge sensors make sense.

What Tom and Bitt said.

Nobody thought those Flip video cameras were worth a damn when they came out, and they sold gazillions of them.

Sorry, Cris. I think you're the only guy I've met who's name doesn't have the "h" in it -- I don't mean to misspell it on purpose.

But please don't misquote me or continue to resort to name-calling. Simply constructing a logical representation of your side of the debate is sufficient.

"Misquote" in that I've never said you don't know how to focus a camera. I simply noted that the focus point in that particular image was not where I'd expected it (the eyes), but rather on the clothing near the bottom of the frame. That's probably the iPhone's fault, or maybe the camera shifted between focus point selection and release of the shutter. Dunno.

Cheers

One other useful thing for the Lytro camera technology: crank the resolution up to at least 720p (they're already at standard def resolutions) and you've now got a very interesting video camera that works in low light without requiring autofocus.

Yea, that picture isn't out of focus. Her eyes are in focus I'm afraid. I'm not mis quoting you either, you clearly say that that image is out of focus.

Again no name calling. I didn't call you any names. I can think of a few if that make you feel better ???

Cheers

Cris

I've been hesitatant to post what I'm about to post, but I think it needs to be said...

Am I the only only who believes this discussion and the one in the iPhone 4S thread are getting way out of hand?

I've been a member of this forum for over a decade now, and this has always been one of the friendliest, most informative places I've known on the internet, filled with nice and knowledgeable people, some of whom I've met in person at a couple of the meets. I'd hate to see this change and it saddens me to see the atmosphere here getting spoilt over what is essentially a simple disagreement. If you don't agree, which you are obviously completely entitled to, can't you guys just agree to disagree and shake hands? Please?

Cris, this post is not meant directly at you just because I've responded to your post. It's a general posting.

Cheers!

Not taken as such. I hear what you are saying.

I think for too long the BBS has been too tolerant in places, it's natural to have disagreements, I'm fine with it. I believe that left unchallenged the discussions lead to indifference of opinion and those who may disagree tend not to post as they feel they will be out cast by the group. So instead they out cast them selves by not posting and eventually never coming back.

This BBS is not the thriving interesting place it used to be. I feel strongly that this is due to the interesting people with interesting opinions getting fed up for the tolerance of a few people who seem, on some levels, just out to enrage people.

The one thing I can tell you for fact is, if any of these two discussions had been held face to face I would have not held out so long to call bullshit and use stronger terms. People who know me in person will know this to be true.

I don't really see what is wrong with what has been said, or the tone in which has been said. We are all big boys and should be able to take a bit of banter. If someone calls my work, which I feel passionately about, I will call them back and make no apology for it. I think that is fair ??? It's not all one way traffic.

I would agree that continued mud slinging does not serve the community well, but on points that people feel strongly about freedom should be given to express that view. I have not stooped to name calling as some would suggest, I have just called a spade a spade and said things as I saw them. I never got personal or made threats, so I can't really see the problem to be honest.

Cheers

Cris