Stories
Slash Boxes
Comments

SoylentNews is people

posted by martyb on Friday October 28 2016, @12:22AM   Printer-friendly [Skip to comment(s)]
from the who-would-want-a-camera-that-was-heavy-and-dark? dept.

The creator of the new Light digital camera explains how he made it work:

The best digital cameras today are SLRs (single-lens reflex cameras), which use a movable mirror to guide the same light rays that fall on the sensor into the viewfinder. These cameras normally have precisely ground glass lenses and large, high-quality image sensors. In the right hands, they can shoot amazing pictures, with brilliant colours and pleasing lighting effects, often showing a crisply focused subject and an aesthetically blurred background.

But these cameras are big, heavy, and expensive: A good digital SLR (DSLR) with a decent set of lenses—including a standard 50 mm, a wide angle, and a telephoto, for example—can easily set you back thousands of dollars.

So most photos today aren't being shot with DSLRs but with the tiny camera modules built into mobile phones. Nobody pretends these pictures match the quality of a photograph taken by a good DSLR; they tend to be grainy, and the camera allows very little artistic control. But smartphone cameras certainly are easy to carry around.

Can't we have it both ways ? Couldn't a high-quality yet still-tiny camera somehow be fit into a mobile device ?

The Light camera starts with a collection of inexpensive plastic-lens camera modules and mechanically driven mirrors. We put them in a device that runs the standard Android operating system along with some smart algorithms. The result is a camera that can do just about everything a DSLR can—and one thing it can't: fit in your pocket.

http://spectrum.ieee.org/consumer-electronics/gadgets/inside-the-development-of-light-the-tiny-digital-camera-that-outperforms-dslrs


Original Submission

Related Stories

Google Reportedly Acquires Lytro, Which Made Refocusable Light Field Cameras 14 comments

Google is reportedly acquiring Lytro, a company that made light field cameras and hoped to pivot to virtual reality video capture. Google appears to have gotten a good (or at least cheap) deal:

Multiple sources tell us that Google is acquiring Lytro, the imaging startup that began as a ground-breaking camera company for consumers before pivoting to use its depth-data, light-field technology in VR.

One source described the deal as an "asset sale" with Lytro going for no more than $40 million. Another source said the price was even lower: $25 million. A third source tells us that not all employees are coming over with the company's technology: some have already received severance and parted ways with the company, and others have simply left. Assets would presumably also include Lytro's 59 patents related to light-field and other digital imaging technology.

The sale would be far from a big win for Lytro and its backers. The startup has raised just over $200 million in funding and was valued at around $360 million after its last round in 2017, according to data from PitchBook.

Despite a lot of hype, Lytro had little success with its expensive, ergonomically challenged, and low resolution light field cameras for consumers.

Also at 9to5Google and Engadget.

Related: LinkedIn's Top 10 Silicon Valley Startups for 'Talent Brand' - Note: Both Lytro and Theranos are on the list.
A Pocket Camera with Many Eyes - Inside the Development of Light


Original Submission

Xiaomi Teams Up with Light to Develop Multi-Module Cameras for Smartphones 7 comments

Xiaomi Teams Up with Light for Multi-Module Smartphone Cameras

Xiaomi and Light, a computational imaging firm, have announced at Mobile World Congress that the two companies will be working together to develop new multi-module cameras for smartphones. The two companies promised that the jointly-developed cameras will feature DSLR-level capabilities, but did not disclose when the first product from the joint project is expected to come to fruition.

Light specializes on computational imaging solutions using multiple camera arrays. The company has gone so far as to develop their own chip that can work with 6, 12, or 18-camera arrays. And while Xiaomi and Light aren't specifying just how big of a camera array they're looking to develop, we're likely looking at something in the lower-bounds of those number, if only due to the limited size of smartphones. For reference's sake, a 6-module camera would be very similar to what Nokia has done for their Nokia 9 PureView.

Cover the entire back of a smartphone with cameras, then gingerly hold it using the corners.

Related: Meta-Lens Works in the Visible Spectrum, Sees Smaller Than a Wavelength of Light
A Pocket Camera with Many Eyes - Inside the Development of Light
Caltech Replaces Lenses With Ultra-Thin Optical Phased Array
Nokia (HMD Global) Partners with Zeiss for Optics Capabilities
Google Reportedly Acquires Lytro, Which Made Refocusable Light Field Cameras
LG's V40 Smartphone Could Include Five Cameras
Leaked Image Shows Nokia-Branded Smartphone With Five Rear Cameras


Original Submission

This discussion has been archived. No new comments can be posted.
Display Options Threshold/Breakthrough Mark All as Read Mark All as Unread
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
  • (Score: -1, Troll) by Anonymous Coward on Friday October 28 2016, @02:02AM

    by Anonymous Coward on Friday October 28 2016, @02:02AM (#419691)

    Don't want. This is a camera that will report back to google where you are and what you are taking pictures of and anything else google can hoover up.

    • (Score: 0) by Anonymous Coward on Friday October 28 2016, @10:38AM

      by Anonymous Coward on Friday October 28 2016, @10:38AM (#419788)
      Only if it has Google Play Services and everything else. If this is supposed to be a just a camera, then they might just be using Android as an embedded OS. The AOSP core of Android has nothing that inherently ties it to Google. The Kindle Fire is an example of an Android device with no ties to Google.
  • (Score: 2) by PartTimeZombie on Friday October 28 2016, @02:17AM

    by PartTimeZombie (4827) on Friday October 28 2016, @02:17AM (#419692)

    I'm a keen hobby photographer, and use my DSLR every weekend ( at least).
    If I had a dollar for every "new technology" that was going to take over from DSLRs I'd have about $8.
    This looks like all of them, kind of interesting in a new tech sort of way, but it's still just a phone camera.
    Also there's a quote on the site from FT.com (as if they would know anything about cameras) that says:

    The L16, if it is as good as promised, will be as important for photography as the first 35mm film camera, the Leica 1, was in 1925."

    To which the answer is:

    It won't be and it won't be

    • (Score: 5, Interesting) by edIII on Friday October 28 2016, @02:34AM

      by edIII (791) on Friday October 28 2016, @02:34AM (#419697)

      Not so sure. I carefully read the whole thing, and the approach is amazing. I would even say this is an entirely new class of camera.

      To take pictures at 70 mm, we move the mirrors so that the five 70-mm modules now point straight out from the camera; all of them cover approximately the same field of view but from slightly different perspectives. We now enlist the 150-mm modules as well, adjusting their mirrors so that they capture four images that tile the 70-mm modules’ field of view. And once again, we combine the many images digitally to provide a better picture than a single-module camera could possibly take, one that rivals a DSLR image.

      I can imagine quite a few hatchet jobs being rushed out. If this thing is affordable for the masses, we will be able to see for ourselves how good it is.

      UNLIKE many other fluff pieces, this one is extremely technical:

      We can even use our technology to zoom anywhere in the range of 28 to 150 mm. Traditional zoom lenses change focal length by physically moving the lens elements with respect to one another when you rotate the zoom-control ring. Our modules are too small to have either the space or the mechanical precision to accomplish this synchronously across multiple camera modules. So we took a systems approach to solving the problem, using fixed-focal-length lenses.

      Suppose you wanted to capture an image with a 50-mm field of view, smaller than what a 28-mm lens captures but larger than that of a 70-mm lens. We activate all the 28-mm camera modules and crop each of the images to the 50-mm frame. (Cropping is not ideal because we lose some sensor area and light.) We also simultaneously use the 70-mm modules. But before we do that, we move the mirrors on four of the 70-mm modules so the captured 70-mm images overlap enough to cover only the 50-mm frame. This way, we retain all of the light collected by the 70-mm modules.

      Well that explains why it can replace a high dollar DSLR, and how it can it zoom. It's one huge image to begin with carefully reconstructed by software

      With all the modules working together it provides a depth map for software to provide even further artistic effects, one of them not easy or cheap to do, and this does all of them that are possible:

      The software can also change what is called the bokeh, which refers to the aesthetic quality of the blur that appears in the out-of-focus parts of an image. Traditional cameras adjust their lens apertures by opening and closing an iris of sorts made of plastic leaves, overlapped to try to mimic a circular opening. As a result, small, bright, out-of-focus objects appear as regular polygons or circular disks. This is the effect most people are used to. Many photographers consider the ideal bokeh as having a very gentle roll-off, with no sharp edges defining the circle—a Gaussian blur.

      Photographers will pay a lot of money for a lens with the right bokeh. In our design, the camera uses software to add blur with the right bokeh to those parts of the scene that are outside the selected depth of field. This approach means that users can get whatever bokeh they want. They can choose the conventional disk-shaped bokeh or one with a Gaussian blur. Or they can get creative—for instance, picking a star-shaped bokeh for use in holiday photos, making small decorative lights appear as stars.

      Just, perhaps, they may have a shot at the title. Ohh, and it's an equivalent 52 megapixels.

      --
      Technically, lunchtime is at any moment. It's just a wave function.
      • (Score: 2) by Zinho on Friday October 28 2016, @03:58AM

        by Zinho (759) on Friday October 28 2016, @03:58AM (#419709)

        If this thing is affordable for the masses, we will be able to see for ourselves how good it is.

        I can get behind that. Unfortunately, from TFA:

        Our first-generation L16 camera will start reaching consumers early next year, for an initial retail price of $1,699.

        So, it's still vaporware, and will be inaccessibly priced when released. He's obviously trying to recover his R&D expenses, since the article itself asserts that the hardware itself will only cost ~$64; the rest is just software. His margins will have to come down a long way before it will get traction in the smartphone market.

        --
        "Space Exploration is not endless circles in low earth orbit." -Buzz Aldrin
        • (Score: 3, Insightful) by Immerman on Friday October 28 2016, @04:31AM

          by Immerman (3985) on Friday October 28 2016, @04:31AM (#419719)

          Well, $64 for the cameras and lenses - the actuated mirror system probably adds at last several bucks to that. And then there's the LCD display, and fast enough processor and RAM to capture, analyze, and merge between 4 and 16 simultaneously taken 13MPixel images into a single final image in near real time. Plus all the electronics, buttons, chassis, patent-encumbered memory slots and file systems, etc.

          I doubt it costs anywhere close to $1700, or even $800, to build the thing, but wouldn't be overly surprised at $200, maybe even $400. And there's the design, tooling, etc. to consider as well - none of that is cheap and it's all a sunk cost before you've made even one unit. While the cost essentially disappears when spread across millions of units sold, if you're only betting on selling a few thousand of your "prototype" product, those costs are thousands of times greater per unit, and you need to plan on recouping that cost if you're going to have a remotely viable business plan (unless you have enough cash reserves to afford to sell the first model or two at below cost as "advertising" for later models)

          Also, I didn't see anything to suggest this is targetting the smartphone market directly, though there's obvious long-term potential for the technology there if it catches on. For starters, I didn't see anything suggesting it's a phone. Seems instead to be designed specifically as a camera, and one designed specifically for photographers who already own thousands of dollars worth of bulky cameras and lenses that are often left unused in favor of the convenience of crappy cell phone cameras. Even if it's inferior to a "proper" high-resolution DSLR camera in terms of image quality, it still has the potential to more than make up for it in convenience, not to mention the ability to tweak many parameters in post-production in ways completely impossible with traditional digital photography.

          Bottom line, it sounds like it has a lot of potential (we'll see how well it delivers), and from what I recall as a non-photographer, $1700 is barely past entry-level pricing for a decent "pro-sumer" camera.

        • (Score: 2) by edIII on Friday October 28 2016, @04:31AM

          by edIII (791) on Friday October 28 2016, @04:31AM (#419721)

          Good point, but at that price point we may find at least enough buyers to verify some of the claims. If you had a major photography magazine for instance, this guy will probably send it there.

          Let's wait and see. On paper it looks good, and it didn't clock in at some insane price point like $50k or something.

          --
          Technically, lunchtime is at any moment. It's just a wave function.
      • (Score: 0) by Anonymous Coward on Friday October 28 2016, @11:37AM

        by Anonymous Coward on Friday October 28 2016, @11:37AM (#419801)

        Not that new: https://en.wikipedia.org/wiki/Lytro [wikipedia.org]

        And that's not including those creatures with compound eyes :).

  • (Score: 3, Interesting) by RedBear on Friday October 28 2016, @08:12AM

    by RedBear (1734) Subscriber Badge on Friday October 28 2016, @08:12AM (#419768)

    For those who aren't already aware, Apple recently bought a company (LinX Imaging) that was making a multi-lens camera system for mobile devices. The system had the capability to take input from lenses with different focal lengths and combine the data into a clearer, brighter, or higher resolution image, or simulate optical zoom capabilities through software merging of the images from the different fixed focal length lenses. I think LinX was demoing dual and triple-lens setups. The technology made it into the Apple iPhone 7 Plus as a dual-lens module, and with the iOS 10.1 update all iPhone 7 Plus owners now have the ability to simulate the kind of nice background blur (known as "bokeh") that you get when you use an expensive large-aperture lens on a DSLR. It uses data from the dual-lens setup to create a 3D depth-map of the image and selectively blur anything not within the chosen focal plane. You end up with a very passable impression of the kind of attractive portraiture effect, the "popping" of the sharply focused subject from the blurred background, that you would normally need a decent DSLR with APS-C or Full Frame size sensor and a large aperture lens to get. It isn't perfect, just what any skilled Photoshop user might be able to do with 20 minutes of careful masking and blurring, but it is pretty darn good and happens instantly right in the camera.

    DSLR fans (or as they like to call themselves, "real photographers") are of course still scoffing at this as just a silly gimmick, but this is of course just an initial peek into what software will be able to do using a multi-lens setup. These multi-lens camera systems will end up being able to do things, or at least simulate things, that no DSLR will ever be able to do. And they will end up being much cheaper than DSLRs because the tiny plastic lenses are a tiny fraction of the cost of even low end DSLR lenses.

    One of the primary issues that remains with DSLRs is dynamic range. There have been slight improvements in dynamic range but it is still incredibly easy to blow out highlights in an image that contains both dark shadows and bright areas. Photographers have to pay close attention to the histogram and underexpose images just to avoid blowing the highlights out. Then they take those images and selectively raise the brightness of the shadows in order to bring back detail. This of course has a tendency to introduce a lot of noise in the image. While there has also been some progress in reducing the total amount of noise captured from DSLR sensors, it is still a tedious and delicate balancing process for photographers to create final images that truly resemble what the human eye sees in a high-contrast scene.

    So how can these limitations be surpassed? One answer is image stacking. You take images using different exposures and stack them together. The nature of this process of course requires that nothing moves in the scene in between shots. The same technique is used for night photography. Many long exposures are taken and then stacked together to produce a clear image of the night sky. These techniques are impossible to use on anything that's moving quickly, and normally must be done with the camera on a stable tripod. But what if you had a dozen sensors all recording the same scene in the exact same instant? You wouldn't need to worry about anything moving between shots. Any pixel in the scene that produced a signal wildly different from the signal from other sensors can be discarded as noise, or rectified to be closer to the signal from the other sensors. Thus, more total light is captured for the scene, and noise is filtered out, and different sensors can be set to different sensitivities to capture both dark and bright areas of the image clearly. Unless you could vary the sensitivity of individual pixels on the DSLR sensor, there is no way a DSLR can ever match what a multi-sensor camera might be capable of.

    So go ahead and scoff at this as some kind of gimmick, but I really don't believe that it is. I believe that this camera represents the future of consumer cameras. Except the consumer camera market is almost dead. So what this really represents is the future of smartphone cameras. All consumer cameras will eventually be smartphones, for many reasons very well laid out in this prescient YouTube video [youtube.com] by Tony Northrup. This all seemed fairly obvious to me as soon as I saw the demo from LinX, and the resulting iPhone 7 Plus dual-lens system that eventually came from it. The only thing that surprises me is that someone already has a much more sophisticated version coming out just a few months from now. It's expensive, but it will be popular, and there will be much more affordable versions appearing in short order. Every high-end smartphone will soon have a minimum of two cameras, then three, four, etc. Within a few years the back of most popular smartphones will look very similar to this Light L16 camera. And they will take remarkably good photos in many situations.

    What will DSLRs still be needed for? Well, extremely high-res landscape photography, maybe. Night/astrophotography, maybe. Extreme telephoto wildlife photography, probably. Studio photography, where you completely control the light anyway, so no need for real-time single-shot HDR capability. Sports, especially low-light indoor sports like basketball, or really fast-moving stuff like racing. Maybe. There will still be plenty of things for a long time that a DSLR will do a bit better. But for many of the things that amateurs or even semi-pros use cameras for, DSLRs will soon end up just not being necessary, or even necessarily the better choice.

  • (Score: 2) by wonkey_monkey on Friday October 28 2016, @03:34PM

    by wonkey_monkey (279) on Friday October 28 2016, @03:34PM (#419866) Homepage

    Can't we have it both ways ? Couldn't a high-quality yet still-tiny camera somehow be fit into a mobile device ?

    No, you can't.

    Oh, you can do all kinds of tricks with software and multiple lenses, but they're just tricks. Unless you've got a total lens area that's the same as a DSLR, you just won't get the same amount of light in, and there's nothing you can do about that.

    --
    systemd is Roko's Basilisk
    • (Score: 0) by Anonymous Coward on Friday October 28 2016, @05:23PM

      by Anonymous Coward on Friday October 28 2016, @05:23PM (#419895)

      Unless you've got a total lens area that's the same as a DSLR, you just won't get the same amount of light in, and there's nothing you can do about that.

      Yeah, but if you have the same total lens and sensor areas, distributed among a dozen or more individual lenses and sensors, you're going to have shorter focal lengths for the same field of view -- the whole thing could be the thickness of a shirt-pocket P&S. (And that's before the folded optical paths this particular implementation uses.) It's perhaps wrong to call it "high-quality yet still-tiny", since it's not tiny at all, but it seems reasonable to me that a multi-aperture approach like this could match DSLR performance with substantially reduced overall dimensions.

      Now this camera, despite having 16 lenses (although only "up to 10" are used for any particular shot), has less lens area than a real DSLR, so I agree it definitely won't actually match DSLR performance. But it has way more lens area than a smartphone camera, and more than a similar-thickness P&S -- so I expect it to beat these.

  • (Score: 2) by MichaelDavidCrawford on Friday October 28 2016, @11:59PM

    by MichaelDavidCrawford (2339) Subscriber Badge <mdcrawford@gmail.com> on Friday October 28 2016, @11:59PM (#419995) Homepage Journal

    While I am willing to accept that the best digital is better than my Pentax K1000 - the cheapest analog SLR money can buy - I cannot afford such nice digital cameras.

    I shoot film then have the processing facility scan it onto a CD for me.

    I'd rather shoot slides but I can't find the film.

    --
    Yes I Have No Bananas. [gofundme.com]
    • (Score: 1, Insightful) by Anonymous Coward on Saturday October 29 2016, @02:31AM

      by Anonymous Coward on Saturday October 29 2016, @02:31AM (#420015)

      I shoot both film and digital. Fellow Pentaxian on a shoestring budget, as chance would have it.

      My fav is the Super A (Program A). Try to pick one up if you can, they sometimes go super cheap. Paid 25 for mine. If you like the K1000, you might fall in love with the A. Great electronic shutter, Tv mode with A lenses, small and light. Watch out for light seal/mirror damper degradation.

      In my opinion, digital is not necessarily "better". On the higher end of current-gen full-frame bodies, real resolution has finally left ye olde 35mm film in the dust. Took sensor makers long enough.

      What I prefer about digital is convenience, low-light-capability and in-camera image stabilization. Operating cost, naturally. In the field chasing motion? Bring on the virtual 5fps motor drive and rip through "rolls" like it's 1989!

      Film still has dynamic range and colour fidelity going for it though. If you have good film stock, time for the analogue process and enough light, you can give quite a few DSLRs a run for their money.

      To add some on-topic: synthetic aperture/bokeh may be the future. But hipster reasons aside, there's something deeply engaging about handling a camera with mechanical inputs. Using the camera becomes part of the photographic process in a way that I can't see a touchscreen operated device replicate. Or any device that's too small to be properly held for that matter. Ergonomics do matter and an SLR is close to optimal in that department.