Previous Entry Add to Memories Share Next Entry
Cool gadget of the day: the Lytro camera
device
jducoeur
I got put on to this one by an article in the Economist a couple of months ago (which I only just got around to reading).

The Lytro camera is arguably the biggest rethink of camera technology in centuries. The underlying concept seems to be straightforward, if Very Very Clever. Most of the expense of a good camera is all those moving parts, and much of the difficulty of using one is getting all the moving parts to do exactly the right thing at the right time. The Lytro turns that logic on its head. Since storage and computing power are increasingly ridiculously cheap, why not leverage them? So the camera does without moving lenses, instead capturing the entire "light field" coming in -- basically oversampling the data enormously. Then, when you have time and leisure, you figure out what you want.

Or in other words: shoot the picture first, then focus later.

The website is crazy cool, and does a nice job of illustrating the point. The picture gallary, in particular, is better than a thousand words -- it shows a bunch of pictures, and lets you interactively click on them to refocus on specific objects.

Add to that the fact that, with no lenses to focus, and much less complex electronics, the thing claims to have no shutter lag -- you click, it records the light field, you move on more or less instantly. Between that, the ability to worry about focus later, and the lack of flash (it has an f/2 lens, so claims to deal with low-light well), this thing may be the *perfect* SCA court-photography camera. I gather that the main drawback is that the maximum resolution is pretty low at this point, so you can't really make 8x10s with it. But it's likely plenty good enough for web photos, and that's the vast majority of what I shoot.

It's not quite out yet, and it's not cheap -- but for brand new tech the price is pretty reasonable ($500 for the best version). I confess a real temptation to buy one of these gadgets, despite knowing that in three years it's going to be way better. Even the first versions sound like a remarkably useful advance, that matches my photography needs rather well...

Michael was talking about this a few months ago-- it sounds really neat! (Though my own camera purchasing will be waiting until the "it's out and cheap" phase of things.)

Marsy's been itching for a new camera and I'm very tempted to get this for her. FYI, in looking at the specs, I wouldn't bother with the $500 version. The only real difference between that one and the $400 model 8GB vs. 4GB of memory. So you'd be paying $100 extra for on-camera photo storage of 750 pictures instead of 350 pictures. $100 seems like a heck of a lot for 4GB unless you think you're going to frequently be taking more than 350 pictures a day. OTOH, the 8GB model comes in red, and that's cool. :-)




...

This comes scarily close to making the Enhance Button plausible. (Warning: TVTropes link.)

Yep. But even more than that, it suggests that the future may involve a serious rethink of what we mean by "photo". We've been assuming WYSIWYG pretty much from the beginning, but one of the interesting subtleties here is that Lytro doesn't even really encourage you to reify your picture into a final version for sharing. Instead, the descriptions sound like you should instead share the *raw* version, and let people decide for themselves what to focus on.

That's both brilliant and logical. But it takes photographs into the land of interactivity in a way that I've rarely even seen in science fiction. Not only does the Enhance Button become possible, it becomes *normal*. And follow that logic into video -- moreover, into 3D video -- and a lot of SF ideas start suddenly seeming a lot more immediately plausible.

Once in a while, you come across an invention that looks more game-changing the longer you stare at it. This thing might well be one of those...

I didn't want to be the first to point out that this could also change how we do public surveillance cameras.

Yaas. Indeed, if this works as billed, the only reason I can come up with *not* to use it is that it is fairly expensive in terms of storage. But given the way that's been getting cheaper, I honestly suspect that any responsible surveillance system can and should refit with these relatively soon.

Now that you mention it, I wonder if they're pursuing that angle specifically. As a cool consumer toy, it'll penetrate gradually. But get the right ear of law enforcement, and they could probably wind up with orders for hundreds of thousands of them pretty fast. Whether this is good or bad I leave as an exercise for the reader...

"Whether this is good or bad I leave as an exercise for the reader..."

The answer is, of course, "Yes."

Not really. 'enhance' usually implies increased resolution, not improved focus. Resolution is limited by sensor size (and pixels) and quality of optics. Focus is pretty much a solved problem if you are willing to choose between fixed focus (requires more light) or auto focus (requires picking a subject). Most phone cameras have no method of focusing, and pictures are acceptable for many people.

What this does offer is out-of-focus options. With a f/2.0 lens the depth of field will be fairly shallow, leaving many things out of focus. You will then have the ability to change which things are out of focus. Being a photographer more into large format cameras, which require careful thoughts beforehand on exactly which things want to be in focus, and which out, I don't see the advantage to changing my mind (or letting others change it for me).

How do they reconcile an '8x optical zoom' with only two controls (power and shutter)?

Honestly not sure. Might be a fixed permanent zoom, or the marketing spiel might be oversimplifying...

A 'fixed permanent zoom' is a 'telephoto'. 'zoom' implies changing focal length. It also adds back a bunch of those moving parts you were happy to have go away. And you don't want one; there are many reasons people use 'normal' lens most of the time, not telephotos.

My understanding, and I'm not a physicist or optical technician, is that by capturing the light field, they are encoding the directionality of light rays, rather than just the 2D landing. So they can reconstruct those vectors in post processing. Zooming is no longer the 'digital zooming' of pixel resizing, but actually reconstructing the vectors entering the camera, and then remodeling those rays through a different mathematical 'lens'. Performing through functions the same process that happens when you change focal length and all the other things that go into a true optical zoom. But it's probably misleading for them to use the term optical zoom in this case.

The "details" section lists:

Controls: Power button; Shutter button; Zoom slider; Touchscreen.

So, actual zoom (which is consistent with an in-depth article I read recently.)

I've been lusting after this since I first heard of it. I haven't been on the cutting edge of any techie thing for a while, so I think I'm going to go for it. It'll be interesting to see how it does in low light. While I love the no-focus nature of it, the thing I've always wanted is something with expanded dynamic range by default, so you don't have to choose whether to shoot the dimly lit room or the brightly lit window.

Is this purported to have a larger dynamic range? I would assume the opposite given the extra work they are adding to the sensor.

I keep hoping that some camera company decides that the megapixel race is over, and work on dynamic range instead. 6MP is probably sufficient for any casual photographer (i.e. beautiful prints up to 8x12).

No, probably not, unfortunately; they don't mention anything about it. It's just the other feature I'm looking for to get something that is more like an eye than a camera.

Very intereresting reading. thx


I think I saw a writeup of this on Ars. The reason for the low resolution is that it is using existing CCD chips, and so it is sacrificing resolution to be able to store light field information. Their original research work used multiple angles and multiple capture devices, but it wasn't technically feasible to replicate that in a commercial product. And they're right in figuring that for the end product of most point-and-shoot photographers, high res is not as big a deal as instant capture and device simplification.

What I like is that it doesn't really need a lot of the things cameras have, like a viewfinder, lens manipulators, etc. You almost literally point and shoot. You're not even holding it up to your face.

I see this leading to higher quality self-portraits and action shots.

Thank you for this. This is... fascinating.

I find the gallery on the webpage both interesting and frustrating. The photos they have for you to play with are too simplistic. It's obvious the places where you should click on the stick the focus. They are not thinking outside the box enough yet, I think.

Like you, I see an application for court photography, not for the silence reasons, although those are good reasons, but for the panoply of people in court. It would allow you to see the entire order assembled for someone's inclusion and then click on everyone's face one at a time to see their expression.

Or battlefield photography: To allow you to truly get a glimpe of what each unit was in the process of doing when the shutter was clicked. Crown tournament photography, for different and similar reasons. (I can see you 3 list fields over, your grace. Perhaps you should have taken that shot?)

It only plays with Macs right now, which is the only reason I have not whipped my credit card out. I am struggling with serious camera WANT.

I took a bunch of photos on the river in Oregon last month. The experience of being in the canyon on the river... the 3d-ness of it just could not be captured on film. Everything looks flat and dead. I bet this camera would have helped with that too.

Might well have, although I suspect some evolution will be necessary to really make that sing.

The neat thing here is the sense that they are really on to something. This first model feels like the original iPod: a bit oversimplified, going to be frustrating in some ways, needs a lot of tweaking and tuning and a whole bunch more power. But I have a strong suspicion that in a few years, once they've evolved both the hardware and software a bit, it's going to change the whole way one thinks about photography.

Specifically on the 3d-ness -- one obvious thing to try (and I'd be surprised if they're not experimenting with it already) is a 3d version of this camera. Two lenses, both recording the complete light field like this -- I suspect there's a whole 'nother doctoral thesis on how you can stitch them together, but I'd bet that you can eventually wind up with a fully immersive image, that looks truly 3d *and* allows you to refocus. It probably could even be tuned to your specific eyes, to work better for many people than most current 3d does.

Then for extra credit, take that idea and combine it with hardware that follows your pupils and refocuses the image based on your current focal distance. Once the software is fast enough to work in perceived realtime (probably 1-2 orders of magnitude faster than their current demonstrations), you can have a "picture" that is immersive beyond anything currently possible -- you can actually "look around" the image intuitively. Now imagine giving a tour through a panoramic 3d image with your eyes, with the image refocusing and zooming based on your eye movements as you show participants around what's in the image. From what I know of the technologies, that seems like it's plausible within five years or so.

I always love it when I get that little spine-tingling sense that a bit of science fiction is about to become real...