Showing posts with label resolution. Show all posts
Showing posts with label resolution. Show all posts

Tuesday, December 6, 2011

LIGHT-FIELD TECH IS A GAME-CHANGER

How the Lytro No-Focus Light Field Camera Changes Photography

 

Building upon years of research by Stanford University, and with a PhD thesis and five further years of research in hand, Ren Ng has finally launched the infinite-focus, light field photography company Lytro. Fundamentally, a Lytro camera — a light field camera — does away with the single most important and limiting factor of conventional photography: the lens. Instead of focusing, distorting, and cajoling light onto a 2D digital or film plane, light field cameras capture everything that it sees. With conventional lenses, a specific focal point must be chosen — with Lytro, you simply capture every last grain of light data, and then use software to define a focal point.

This is quite a hard concept to get your head around, but it helps if you think of conventional photography as the equivalent of a flat, 2D bitmap or raster graphics, while light field photography is like 3D vector graphics. With conventional photography, what you see is what you get — you can’t resize your photos without losing data, and you certainly can’t alter the focal point. With light field photography, every single ray of light is recorded in a lossless fashion. You then create a virtual focusing point, and ray tracing is used to compute a digital bitmap image. Raw light field photographs will understandably be quite large — but not prohibitively so, according to Lytro.


With the technology out of the way, we should now focus on how light field photography will change the very fundamentals of photography. The basis of photography involves a hole — a lens — and some kind of 2D plane; a wall, a CMOS sensor, light-sensitive film, and so on. The quality of the image produced is entirely dependent on the lens that you use. With light field photography… there is no lens. Or rather, there is no lens in the conventional sense of the word. For over a hundred years we have struggled to focus the 3D world onto a 2D plane — and now, with one fell swoop, we can do away with heavy glass and entirely remove the concept of apertures. With light field photography, you point your camera at a scene and take a photo. No focusing, no thinking, just snap.

No more will candid low-light photography be limited to photographers with enough money to buy a $10,000 camera and lens — and no longer will you have to buy the perfect-bokeh 85mm lens for portraiture. Just point your camera at the subject, capture some light, and then play with the photo after the fact. Lytro will further democratize the digital photography field, allowing anyone and everyone to take perfect photos every time. The only difference between pros and amateurs will be the direction in which you point the camera — and the lengths you go to to find the perfect landscape, of course.

At the moment it sounds like the first Lytro product will be a consumer-grade point-and-shoot affair, which means you’ll probably have to use some kind of “focus browser” on your desktop or laptop to produce final images. Assuming light field technology can be scaled down to smartphones, though, it’s safe to assume that we’ll be able to pick a focal point using an app.
As for when the first Lytro camera will appear, we’re told that it could be on the market before the end of the year. There’s no data about how much it will cost — and don’t be surprised if the first light field camera costs a lot.

Read more about light field photography, and play with some Lytro photos

Friday, December 2, 2011

IPHONE VS. DIGITAL CAMERA!? SAY WHAT??

I'm always wondering if my bulky equipment will soon be replaced with a pocket-sized camera that will do the same thing. I'm sure it will happen the moment I upgrade to the giant $5000 pro camera with all the bells and whistles. Here's an interesting article I found on a really cool website called ars technica.

 

Can the iPhone 4S replace a "real" digital camera?



Can the iPhone 4S replace a "real" digital camera? Ars investigates
Clockwise from top left: Olympus XZ-1 point and shoot, Canon 20D DSLR with EF-S 10-22mm lens, iPhone 4S, Samsung Galaxy SII. Photo taken with an iPhone 4.

When Apple announced the iPhone 4S, the company certainly talked up the improvements made to the smartphone's integrated camera hardware. With 8 megapixels of resolution, a redesigned lens, and enhancements to its software, it certainly meets or exceeds the needs of most casual shooters.

Still, we here at Ars have received plenty of questions to the effect of, "can the iPhone 4S replace a 'real' camera?" That's actually a hard question to answer, because individual needs vary widely. Would a professional photographer replace her trusty DSLR with an iPhone 4S? No. But, might a casual snap shooter replace a pocket camera with an iPhone 4S? It's pretty likely.
Making a decision to ditch your camera for a smartphone is always going to be a matter of personal preference and a trade-off between convenience versus quality. With that in mind, we set out with a stable of cameras to give as fair a comparison as possible. Along with an iPhone 4S, we also took sample shots with a 5MP iPhone 4, an 8MP Samsung Galaxy SII, an 8MP Canon 20D, and a 10MP Olympus XZ-1.

There are two things we did not do, however. For one, we didn't compare every generation of iPhone. Camera+ developer Lisa Bettany already put together a really thorough comparison already, and suffice to say if you have anything older than an iPhone 4, be prepared to be blown away with the results from an iPhone 4S. If you already have an iPhone 4, the improvements to image quality are a little more subtle. But as we noted in our iPhone 4S review, the speed of the camera hardware and software is significantly improved; this alone is a significant factor that improves usability.

A photo taken with each generation of iPhone, from the original 2MP fixed focus iPhone to the 8MP autofocus iPhone 4S.
A photo taken with each generation of iPhone, from the original 2MP fixed focus iPhone to the 8MP autofocus iPhone 4S.

The other thing we didn't do was compare the video capabilities of the iPhone 4S. We already found the iPhone 4's 720p recording ability to be sufficient for casual use, and at least one filmmaker managed to shoot an entire film using iPhone 4s. The iPhone 4S steps that up to 1080p resolution, and some curious filmmakers at Robino Films found that in static, well-lit scenes, it compares favorably to a Canon 5D mkII. Rolling shutter effects will be a problem, and lowlight shooting introduces some artifacts, but overall the hardware performs well shooting video.

iPhone 4S / Canon 5D mkII Side by Side Comparison

What we were interested in was evaluating the iPhone 4S as a still camera that could possibly replace a 'real' digital camera. To try and keep things as fair as possible, our Canon 20D was outfitted with a Canon 10-22mm EF-S lens was set to 22mm (a roughly 35mm equivalent) and left at f/4.5. Exposure was set automatically using aperture priority, just as most smartphones do. ISO was set manually (the 20D doesn't have auto ISO available in aperture priority mode) but was set to our best guess for the situation at hand.

Likewise, the Olympus was put into aperture priority mode, with its aperture set to f/2.5—similar to the smartphones we compared against. ISO was set automatically, and the lens was left all the way at the wide end—roughly 28mm focal length equivalent.

We took a series of subjects in a couple different lighting scenarios: flowers in outdoor lighting, a building at sunset, and some shots in very low indoor lighting at home and at a bar. We'll show you our comparison shots, noting the performance in each case as well as considering what each device was like to carry, pull out, set up, and take each shot.

Outdoors


We took our band of cameras outside during a brisk autumn afternoon with a clear blue sky to capture some images in bright light. First up is a series of photos taken of a flower arrangement.
Left: iPhone 4S, 1/3500 f/2.4 ISO80. Right: iPhone 4, 1/1150 f/2.8 ISO64.
Left: iPhone 4S, 1/3500 f/2.4 ISO80. Right: iPhone 4, 1/1150 f/2.8 ISO64.

Left: iPhone 4S. Right: Samsung Galaxy SII, 1/1400 f/2.6 ISO32
Left: iPhone 4S. Right: Samsung Galaxy SII, 1/1400 f/2.6 ISO32.

Left: iPhone 4S. Right: Olympus XZ-1, 1/2000 f/2.5 ISO100
Left: iPhone 4S. Right: Olympus XZ-1, 1/2000 f/2.5 ISO100.

Left: iPhone 4S. Right: Canon 20D, 1/1000 f/4.5 ISO100
Left: iPhone 4S. Right: Canon 20D, 1/1000 f/4.5 ISO100.

Here you can see the cameras capturing lots of color. Directly comparing the iPhone 4S to the iPhone 4 in the first image, there are two important factors to notice about the improved camera: a greater dynamic range and a tendency to keep colors—especially reds—from over-saturating.

Apple claims to have improved white balance in the 4S, and as such, its images tend to be a little less cold in regular lighting; they show a little less yellow (as we'll see later) under indoor tungsten lighting compared to the iPhone 4. It's difficult to say for sure why these improvements couldn't be passed on to the iPhone 4 with software, but it's likely that the white balance improvements are at least partially due to the new 8MP sensor and its hybrid IR filter.

Our relatively high-end point-and-shoot, the Olympus XZ-1, over-exposed the image quite a bit. It's hard to say what tripped up its exposure system, but suffice it to say, having a more expensive "real" camera doesn't automatically guarantee better results. The XZ-1 has the benefit of offering a variety of manual controls, including ISO adjustments, exposure compensation, and manual exposure control that would let us dial in a better exposure if that's what we wanted.

The iPhone 4S, on the other hand, has no such control. Tapping around different areas of the viewfinder will bias the auto-exposure system for that particular part of the image, however, so there is a way to slightly override it. Still, we found that iPhone would generally nail the exposure about nine times out of ten without intervention.

At screen sizes, there really isn't much difference between the iPhone 4S, the Galaxy SII, and the Canon 20D. We note that the exposure and dynamic range of the Galaxy SII is more like the iPhone 4, while the Canon 20D's rendering is more like that of the iPhone 4S.

100 percent crop of flower images. From left to right: iPhone 4S, iPhone 4, Samsung Galaxy SII, Olympus XZ-1, Canon 20D.
100 percent crop of flower images. From left to right: iPhone 4S, iPhone 4, Samsung Galaxy SII, Olympus XZ-1, Canon 20D.

Looking closer at 100 percent crops of each image—pixel peepers rejoice!—the differences are still subtle, but the iPhone 4S and iPhone 4 are clearly noisier than the Olympus or the Canon. The Galaxy SII shows signs of aggressive noise reduction that leaves odd artifacts, though at screen sizes or in smaller prints, these won't likely be noticeable. Both the iPhone 4S and Canon 20D show a little blurriness, but given the high winds on that particular day, I attribute this to a bit of motion blur, as other details in the image are sharp.

Photography and Article by Chris Foresman