“Shoot now, focus later”

I know I should probably have an opinion about the recently announced (but not yet released) digital camera by Lytro, which takes “living pictures” and promises a “Picture Revolution.” So, let me form one here as I still ponder how to interpret this Picture Revolution within the historical context of photographic technology since the early nineteenth century.

If you haven’t heard, Lytro (kind of a mash-up of “light” and “nitro”) has claimed to have developed a unique and sophisticated way to capture far more light information than your run-of-the-mill 25 megapixel DSLR. It does this, the website explains, by tapping into and capturing mega-information of the light field, which the Lytro folks define as “the amount of light traveling in every direction through every point in space – it’s all the light rays in a scene.” In short, it’s vector information of light rays that is added into the equation of a light-field camera photograph. The upshot of the technology is that with the info gathered, you can (in-camera, I presume, or else it wouldn’t be much different from post-processing on a computer), “fix” or creatively change the field of focus. I encourage you to read Lytro’s layperson’s explanation of the technology; it’s interesting. The examples (simulations?) that you can play with on the Lytro website can be compared to on-the-fly rack focusing. But apparently you can turn a 2D photo into 3D via software that can manipulate the gathered light information. The sales pitch for this “light field camera” is two-pronged: to correct a missed focus and to allow for “creative” post-processing.

Before I lay into any kind of critique of the idea (can’t really critique the device yet), I’d like to say that the idea of innovating how light is captured and processed in this way is impressive. There’s real science behind it; this isn’t some bogus toy. The brainchild of Lytro is Ren Ng, whose 2006 Stanford dissertation “Digital Light Field Photography” is the basis for Ng’s 2006 start-up company that has brought to fruition this invention. I downloaded and started reading the dissertation — just the intro so far; I wonder how much I’ll be able to understand beyond that — and I’m truly impressed by the thought behind the idea. But I think that ultimately I disagree with its rationale and might not like its implications. Here’s the one-page overview in full:

This dissertation introduces a new approach to everyday photography, which solves the long-standing problems related to focusing images accurately. The root of these problems is missing information. It turns out that conventional photographs tell us rather little about the light passing through the lens. In particular, they do not record the amount of light traveling along individual rays that contribute to the image. They tell us only the sum total of light rays striking each point in the image. To make an analogy with a music-recording studio, taking a conventional photograph is like recording all the musicians playing together, rather than recording each instrument on a separate audio track.

In this dissertation, we will go after the missing information. With micron-scale changes to its optics and sensor, we can enhance a conventional camera so that it measures the light along each individual ray flowing into the image sensor. In other words, the enhanced camera samples the total geometric distribution of light passing through the lens in a single exposure. The price we will pay is collecting much more data than a regular photograph. However, I hope to convince you that the price is a very fair one for a solution to a problem as pervasive and long-lived as photographic focus. In photography, as in recording music, it is wise practice to save as much of the source data as you can.

Of course simply recording the light rays in the camera is not a complete solution to the focus problem. The other ingredient is computation. The idea is to re-sort the recorded light rays to where they should ideally have terminated, to simulate the flow of rays through the virtual optics of an idealized camera into the pixels of an idealized output photograph.

Ng is right — it’s all a question of information. Where he is wrong — or, rather, where he presumes — is in his description of “the focus problem.”

First, I object to the way he constitutes “the focus problem” as “pervasive” and “long-lived” as if it’s simply an inherent problem with cameras and lenses — they can’t be focused properly and we’ve been struggling to focus them since the beginning of time. Poppycock. Obviously whether you’re going for sharp or soft focus or  selective focus, a photographer with a modicum of skill can hit that focus. In fact, with today’s sophisticated digital auto-focusing system, it’s hard NOT to hit the “idealized output photograph” even if you have no skill. I can put my DLSR on full auto and blindly point at a scene and 9 times out of 10 it will be “accurately” focused.  In other words, “focusing images accurately” is not the problem Ng is making it out to be. His framing of the problem is misleading if not bogus; it’s more marketing ploy than anything else.

Second, the language of that last sentence is too restrictive for my tastes. Unless Ng is allowing for the user to decide where light rays “should ideally have terminated” and what is “an idealized output photograph” in an “idealized camera” then I’m afraid he’s defining a photograph that a robot could take. It’s not so much the old debate between “straight” and “artistic” photographer that’s implicitly signaled here that bugs me, but rather the naturalized presumption that there exists a definable “idealized output photograph.” In other words, what I protest are the conditions in which such a phrase as “idealized output photograph” could ever be imagined and uttered. A photograph as “idealized output”? Yikes!

But, I will point out that the conditions giving rise to such an utterance has a long trajectory that does indeed come out of the straight versus artistic debate. Ng’s view and his light-field camera are the logical digital extension of the pursuit of the most perfectly focused and tonally rich (or panchromatically perfect) photograph technologically possible. Although it would be difficult to imagine the Group f/64 types endorsing the kind of post-processing and manipulation of an image that Lytro makes possible, they share a similar attitude about the photographic ideal, one that I think ultimately encloses rather than opens photographic vistas. Ng, I’m sure, would object to the notion that Lytro shuts down photographic vision and argue on the contrary that the user now has even more creative control over the image.

I would counter that any such control is bound within the parameters of the software and the assumptions about photographic ideals that are built into it. So, with Lytro I could rack focus after the fact and turn 2D into 3D photos. I see that as rather limited, even boring once the novelty wore off. I can already see “creative” but cheesy Lytro shots posted on Flickr by the zillions. Everyone will be “clever” in playing with depth of field on the same shot. And there is sure to be some cutting-edge “digital new media artists” who will apply the technology in some dazzling avant-alt.net way to win a MacArthur Foundation Genius Grant and go on to captivate the masses and the critics with some interactive immersive-experience extravaganza in stadium-sized shows set to Philip Glass music and somehow involving laser lights and mirrors and acreages of nanotech fabric on loan from Christo. And I don’t even like 3D.

My bigger concern about Lytro — and let me repeat that I am impressed and fascinated on an intellectual level by the technology; I think it’s brilliant — is that it’s another step into the dreary morass of sloppy, thoughtless photography. Digital already encourages the carpet-bombing approach to taking photographs (I know because I happily did it with my Canon 40D under my “10% rule” — if at least 10 of 100 shots turn out good, that’s success.) There’s no real cost to shooting tons of digital (no film and processing costs) and with auto on it’s fast and easy compared to manual film cameras, so you can afford to carpet bomb. But just as fast food erodes culinary sensibilities in the name of speed and convenience, I think auto-digicams erode photographic sensibilities. You are less apt to take the time to really think about what you’re doing. I think the time and money invested in film photography by mere economic considerations breeds more careful and thoughtful photography. Lytro, I fear, will further pave the path down photographic thoughtlessness even as it is being promoted as being able to capture the “all the stories in a single snap” of a scene that a “conventional camera” misses:

(note that this video is marked “simulation”)

The counterargument to my objection would be that the thoughtfulness is simply displaced to a later moment when you might actually have more time to be more thoughtful with the creation of the image. Perhaps. But then you have really played fast and loose with “capturing the moment” and “capturing space” (see “The Decisive Moment?“) to the point that your “Picture Revolution,” Mr. Ng, bears little resemblance to light-writing (photo-graphy). And that’s okay. It becomes another path to imagemaking and an interesting one at that. But I don’t think it solves any “problems” in photography. Photography is doing just fine without light fields, thank you very much….