You're wrong about weight and cost. The iPhone 7 camera costs Apple $26 per unit. An array of ten would only begin to approach what Canon spends making a single 24-70 zoom. A dictionary won't help you here. Time to slither away?
No. If you zoom in, the text around the smallest lenses says "35mm wide angle".OK, let's look at staking.
This is the camera
my guess is that it is about 9x16cm
Take a look of where the lenses are.
5x 28mm
Not what you think will happen. Stacking and stitching are done all the time today. This isn't science fiction. Look up the Brenizer Effect. The whole excitement is simply that in the near future it will be available in a small, somewhat affordable package (not this particular camera), with output that could previously only be done with large heavy cameras and lenses.
What do you get when you pan your iPhone around (by hand) in panorama mode? That's stitching (and yes the stereo possibilities are there too, bonus). With stacking, the software will obviously have to correct for the slight offset in perspective. I take it you've never used modern post processing tools? In Snapseed (a free phone app) you can now change the direction someone's face is posed, after the fact.That was a mock up of the original idea. Since then the 35mm has become (in theory) a 28mm lens.
Staking is certainly done but not from photos taken from different positions, if I am not mistaken.
Put the equivalent of a 150mm lens on a camera on a tripod. Take a photo.
Now move that lens up /down or sideways to match the position of the one on the L16. Take a photo from each position, then stack them.
Keep in mind that the L16 lenses don't move up/down/sideways so you need to keep the lens parallel to the first shot.
The 150mm lenses are one on top of the other on the left side (the top one is off side to allow for the Light logo...)) the fifth one is about 10cm away on the other side.
Maybe I am not explaining the point very well.
Maybe this will help.
This is a typical (old time) stereo camera.
The distance between the two lenses is about 6cm, approximating the distance between our eyes.
The L16 has lenses of the same type (say the 28mm) further from each other than that.
So my question is, if you took a photo with that stereo camera and moved it a few cm three times (3x 2 photos) and then stacked those photos, what would you get ?
Yes, this is correct. In fact, stacking only offers resolution benefits when there are slight changes to alignment. Ideally, this would be a sub-pixel shift movement, but slight changes in perspective are easily corrected for.
Worried? Hardly. Just because DSLR guys do not buy into every flavor of the month technology doesn't make them Luddites. Maybe if there weren't so many fanboys so eager to see the DSLR demise and are willing to post about it endlessly or buy into every flaky technology, these guys wouldn't appear so defensive. How many threads were there called "DSLRs rule!" were there before the ML and MFT guys started going after DSLRs? Not many that I remember.
Totally agree with this. This L16 is a perfect example. I don't know the specs on the sensor size, but I'd guess it's something like a 1/2" sensor or smaller. If this was the size, then a 4-module camera would capture the same light as a 1" sensor. So that's the maximum level of shot noise performance we could expect--it's a physical constraint.
Or, I suppose they could do 16x 1/2" sensors, each with a 6-17mm F/2.8 lens. Doesn't seem to be what's happening here.
It doesn't have to match the same total light as the larger camera to get similar noise performance. Noise is random. If you take 5 simultaeous photos from the camera modules, and merge them, the differences are removed, ie the noise.
Yes, totally agree. It's cool stuff, and I hope they do well; but they do need to execute and fulfill a need for consumers.
Au contraire, mon frere. I am well aware of this technique and use it frequently. I last used it for a landscape shot just yesterday; or you can see my reply earlier in this thread for this technique when it comes to planetary photography.
Funny thing is that I've been taking photographs for 62 years and have yet to use a "future camera." I use cameras that are available today or were available a few years ago and still produce great images. When the future becomes today then we'll see what's best for a particular use. Until then we're all stuck with today's or yesterday's real cameras.
Neither one of us knows if this particular prototype even uses this processing technique.
I didn't say that it did (though I'd say it's a pretty safe bet, since it was confirmed to have merged all the images, which by definition would blend the noise).
The idea that you could have a 12MP 30mm f/2.2 camera, in a 6mm thick phone computer, that had more processing power (and utility) than the then current laptops, would have seemed like Buck Rogers stuff in 2006. I'm holding that "future camera" in my hand right now.
2006 was not some distant dark-ages past. We all knew technology would improve. In 2006, the iPhone was on the horizon, phones were already thin and powerful, and they all had pretty decent cameras. My 2006 phone had a camera that could rotate for selfies, and produced images that exceed current Instagram posting sizes (and exceed 1080 resolution).
I see clearly that you don't get it. When it is not yet a product that you can buy then you cannot take it out for a spin and know whether it fills your needs. You have only descriptions that are often lacking in sufficient detail. It's fine to think about future camera technology but I'm taking photos with present technology and I'm not waiting for the future to arrive. In my experience the "future" often arrives with multiple changes in realizable technology.
Yes--larger sensors are not just linearly more expensive than smaller sensors. Several small sensors covering the same surface area as a single large sensor are typically much cheaper to produce. This is counter-intuitive, but true and not as preposterous as it sounds.
Well, that's science. Which I suspect is why you don't want to fixate on it.
This is math and science which work the same for all cameras. The trade offs are very well understood. Noise versus resolution versus DOF versus lens size versus exposure time.
I really don't know what your point is. We take photos today with what we have on hand, and we speculate and bicker about what might come in the future in threads like this. Nobody is "waiting".
