A camera you focus AFTER taking the picture.
Lytro.com / Eric Cheng — click the kitteh!
Photography has always required a few technical decisions before shooting. You (or your expert proxy, the camera’s computer) need at the very least, to focus the lens and set the exposure, based on your artistic intent, the subject at hand, its setting, and various parameters related to constraints specific to your lens and sensor or film stock.
Those choices always take a little time, and define irrevocable characteristics of the final image. A seriously out-of-focus subject will always remain a blur, while a blown-out highlight will always glare whitely and detail-less out of the print. Short of repainting the subject, there are limits even to post-processing stalwarts such as Photoshop.
But of course ‘exposure’ and ‘focus’ are concepts of optical engineering, of the process of photography, not characteristics of the world. The visual world is a field of light, from within which field a photograph is constructed by placing the limitations of a specific combination of lens and sensor at a specific location, at a point of time. The visual world itself is neither focussed nor unfocussed, neither over- or under-lit. It is merely light.
What if we could capture more fully a description of that field of light, and, after the fact, at our leisure, decide on what should hold focus, what should be the depth of field, what should be a highlight, what a mid-tone, what deep mystery of shadow? Aesthetically, such a choice offers obvious freedoms. Pragmatically, it means that time need not be wasted on camera-system configuration, but rather in getting access to, and framing the shot –- important for sports, street and wildlife photographers alike.
Two emergent technologies offer to deliver on that promise of leisurely, post-production exposure and focus. On the exposure front, High Dynamic Range sensors have been ‘the next big thing’ for several years now. When it comes to focus, the revolution may have just taken place. US-based startup Lytro has decloaked from stealth mode to announce its first Light Field camera, based on founder and CEO Dr. Ren Ng’s academic research. The camera itself is still vapourware, but Lytro’s algorithms are demonstrated in interactive Flash images on their site. Click on any part of an image, and hey presto, it springs into focus. This is a genuinely disruptive imaging technology, and comes with other interesting claims, including massively improved low-light performance, and single-lens 3-D.
It all looks cool — though rather low-rez — in the demos. Given the potential value of their intellectual property, we’re surprised that Lytro aren’t focussed (ahem) on licensing it out to the big players. Instead, they are (apparently) planning to swim their cool tech, under the flag of a startup brand, into the perilous waters of the consumer electronics market, buoyed only by a vapourware product and a brand positioning — living pictures — which is a weak undersell of the scope of their innovation. One has to wonder, why not aim elsewhere? Why not focus on the professional photography customers to whom Lytro’s very real benefits should have a high commercial value? Why, indeed? Is it possible, for example, their Light Field technology throws away resolution in exchange for its magic? Maybe there’s no way to integrate it with legacy optics, which would reduce its utility for professional photographers? There’s simply not sufficient information on Lytro’s site to tell.
Damn, though, their tech is cool, and double-damn I want one of their cameras, to take for a spin some steamy night at Superstore. We’ve asked them for an interview. We will keep you posted. In the meantime, we’re off to read Dr Ng’s original research papers…
UPDATE — some answers from Lytro…
[More animation 
