Lytro has just changed photography forever
True paradigm shifts don’t happen often, but when they do they have the power to change our entire way of thinking about a given subject.
Today’s story comes from the field of photography.
As any amateur shutter bug will tell you, other than image composition (how you frame up your subjects in the camera’s viewfinder), focus and depth of field are the most powerful tools in a photographer’s creative arsenal.
When you get ready to take a photo, the first thing you need to decide is where your focus will be. For most of us, the habit of depressing the shutter button half-way to initiate the auto-focus before triggering the capture process is second nature: Choose your subject, focus on it, snap!
The combination of focus and the aperture you’ve chosen leads to the effects we’ve all become familiar with: Portraits where the only the subject is in sharp focus and everything else is soft, landscapes where the entire image is in focus letting you see every detail of a mountain or field, and those clever in-between shots where the foreground and background are hazy but the middle distance is sharp.
These creative choices have to be made by the photographer at the time of the shutter-click.
Everything else – colour, contrast, even exposure can be modified after-the-fact in programs like Photoshop.
Until today that is.
Welcome to world, Lytro, a company founded by Stanford University Ph.D. grad, Ren Ng. Ng and his small team have been working quietly for the past few years to develop a technology known as Light Field Capture.
According to Lytro’s website, “the light field is a core concept in imaging science, representing fundamentally more powerful data than in regular photographs.” What this means in practical terms, is that any camera equipped with a light field sensor is capable of capturing much more than colour and intensity which are the two data points acquired by a typical digital camera’s CMOS or CCD sensor. The light field sensor camera “captures the color, intensity and vector direction of the rays of light. [emphasis added]”
When your camera knows this much information about the scene you’ve captured, it can yield astonishing effects – the most immediately thrilling of which is that you can change your focal distance after you take the photo.
It’s a little hard to appreciate what this means until you’ve seen it in action. Thankfully Lytro has provided some sample images for us to play with.
Take a look at the image below. See how the focus is on the toy that the black cat is playing with in the foreground? Okay, now click on the grey cat in the background.
I know, right? It’s kind of mind-blowing. It’s as though we are still in the process of framing up a photo that hasn’t been taken, and we’re adjusting the focus on the camera. It’s not purrfect (sorry couldn’t resist) – the very furthest objects like the couch remain soft even after clicking on them, but you can see the potential this technology has.
When I showed this example to a few colleagues around the office, the reaction was mixed. Some were thrilled, but others bemoaned the technology as a “cheat.” Their premise being that one of the core skills you need to acquire as a photographer is understanding the correct application of focus and depth of field. They’re right. Or at least, they used to be right.
Lytro has fundamentally changed the way we can now approach our images. No longer are their major attributes locked in place when we push the shutter button. They have become dynamic – “Living Pictures”, to use the term that Lytro has adopted. Lytro plans to bring the first light field sensor camera to market later this year.
And like it or not, I suspect that in less than 5 years, every digital camera on the market will be using this technology. You’ll be able to enable or disable it as you see fit, but most people will leave enabled, for the same reason that most people shoot in full colour instead of black and white. You can always switch to black and white later if you want to, and likewise with light field – you can always change your focus later if you want to.
one baby step closer to the technological singularity.. : )
Wow that is super, you can click on the toy, the black cat or the grey cat (but not the window?) and the focal point changes, interesting technology…. I like it
As a avid photographer this technology both excites and saddens me. We are getting farther and farther away from capturing a shot, and more into creating them.
I have never thought of “capturing” a shot but instead, “creating an image on the fly and then adjusting it show show how I feel/think about the image”. I’m a real fan of Photoshop and it will be interesting to see how digital photographers incorporate this new technology into their compositions.
real photographers SHOULDNT encorporate this new technology into their photos. its great for moms and stuff, but theyre are enough people now who claim to be professional photographers who arent.
how many more will just pop up out of nowhere once this technology is introduced…
And there are no moms who ARE professional photographers?
Coming soon to an Iphone 6 :)
This will be especialy fantastic if this technology can be used in macro mode, as when you take a photo of an insect, now all parts of the insect can be focused on. Also will be cool if this tech can make the whole picture focused at the same time. This really is a big step forward.
I like it
A few more buttons on the camera I will not use
but like the idea of being able to.
Now if I could get self focusing eye glasses
I could really take clear photos.
Is this just a web application, or just pc related to show the pictures. How does the concept work for print?
I like the idea of web based, and having the Java or an application make it change the focus points…
I presume that most instances of an image will be rendered with a specific focus. This should allow the photographer to render different finished results, depending on what he/she wants to do.
When I was in military photography, we joked about a mythical magical substance called “fouscene”, which would bring out of focus photos into sharp ones. Now focuscene actually exists and I love it.
“INTERPOLATION”, if my spelling serves me correctly, is what I see here. This is in a way an old method of ‘creating something that’s not there but filling in the missing detail with & based on surrounding information’.
That said, and I agree we are looking at a very grainy sample above (cats), but it is what I expected.
As for the actual ‘capture’ moment, in a sense does this technology actually simply regard each sensor reading as if it had a high F-stop (everything is in focus) and then simply allows the user to select afterward which focus is desired?
“POOP In, POOP Out”! That is the time-proven motto for graphics (& audio). In otherwards, if you start-off with low-res detail, unless you are a very skilled graphic specialist who can actually re-construct such from nothing, you get what I see here, but without the time required to manually interpolate.
Otherwise, if the moment allows, I would just crank the F-Stop, get everything in focus, then sinply Photoshop the desired/required focus effect. This is somewhat ‘interesting’, but I will ‘focus’ my attention on it as it develops further to understand it a little better. Seems like they are actually trying to ‘bust’ that problem of wishing data that’s not there to be there all of a sudden! One can always work downward with resolution, not up (without ‘interpolation’). I am both a graphics & audio specialist, btw. Ho hum.. Next!
Interpolation implies that new data is being inferred from old data, as they sometimes do with getting greater resolution or performing digital zoom. This is different. The sensor actually detects a new, third type of data. I didn’t get into the deep tech in this post, but you should check out their site, they explain it all and I genuinely think this is more than just fancy programming.