Lytro's Allows You to Worry About Focus After You Shoot

[Lee Yuan Sheng]

<div class='os_post_top_link'><a href='http://blog.digitalrev.com/2011/06/22/new-camera-lets-you-shoot-first-focus-later/' target='_blank'>http://blog.digitalrev.com/2011/06/...st-focus-later/</a><br /><br /></div><p><em>"Well, a company by the name of Lytro (sounds a bit like diet supplement or something) has apparently solved all those issues by creating a technology that will let you take the photo first and focus it later on your computer, according to a New York Times article."</em></p><p><iframe frameborder="0" height="500" scrolling="no" src="http://www.lytro.com/pictures/lyt-18/embed?utm_source=Embed&amp;utm_medium=EmbedLink" width="500"></iframe></p><p>Lytro has been making the news with their announcement of a camera that can capture an incredible depth of field first for post-processing depth of field effects later, but part of me wonders whether photographers want to spend more time in front of the computer as it is already? Even for a personal shoot, I easily wind up with 10-20 photos I want to keep, and spending the bare minimum 5 minutes in an image editing program for each means 1-2 hours in front of the computer!</p><p>Talk of technology in DSLRs seems a bit premature; Lytro seems to be targeting consumers with their first camera, and are going it alone without licensing the tech to the usual camera manufacturers. Application in a DSLR might be used to correct back and front focus issues however. Maybe a smart DSLR could keep the data as metadata, and present the option to move the focus distance in a RAW editor. But once again, do photographers want to spend more time in post-processing?</p><p>As for consumer compacts, this might be a way to get depth of field effects without buying a large sensor camera that is likely to be larger and more complicated to use. I wonder if the average consumer will be that interested in such a camera though. Maybe Lytro should just license the tech to mobile phone companies instead; that could well be a real win right there.</p>

[Jason Dunn]

I personally think it's pretty crazy for them to try and compete in the cut-throat world of point and shoot cameras. They're going to fail unless they can come up with a crazy-awesome product that's compelling. This, by itself, isn't good enough. Licensing would make WAY more sense.

From a pure physics point of view though, on a camera if you want deep focus, you'd need to go f/8 or something similarly high...which has an impact on ISO, shutter speed, etc...it seems like a big sacrifice in order to be able to play with focus after the fact.

I was initially impressed by this, but it seems like a big FAIL now. Guess we'll see what they have up their sleeves...

[Lee Yuan Sheng]

I was reading the technical paper by the founder about how "light fields" can be used to assemble a picture, then I realised that the Flash demo allows zooming, and boy, are the photos grainy. Maybe it could well be as you said Jason, that the camera is nothing more than a fancy lens that starts at f/8...

[ptyork]

A few technical points here. First, there are really two lenses at play here. First, a large aperture main lens similar to a regular camera. Second is a micro-lens hovering just above a standard digital sensor. This is most definitely not some trick where they just put an f/8 or smaller lens and add in some artificial blurring to make it look like DOF. It is a totally different optical concept. Very fascinating, actually.

As for the quality of the images in his "technical paper," please note that the technical paper is his doctoral dissertation and that all of the images are from a very crude prototype. If you read it, it talks about how poor the quality is compared to what would be possible using a high-quality microlens array attached to a much higher resolution sensor.

As for spending more time in post-processing, I'm thinking that two extra seconds to select the desired focal point is a small price to pay for not having to focus at the outset. And yes, I expect it to be similar to the RAW workflow where you will export JPEG's, not ship out refocusable flash files or something.

Anyway, this is a technology in its infancy. Give it some time. No, it doesn't look to be there yet, but it really does look to have the potential to revolutionize the field. I mean it's only been five years since this guy defended his dissertation.

Oh, and as for why you would NOT want to partner with a major player...well, do you really think Canon and Nikon are going to run with something that has the potential to kill an entire eco-system? They are being dragged along kicking and screaming into the EVIL world. And EVIL simply merges some of the strengths of the SLR with those of the mirrorless compacts. This would be truly radical, and for these companies, and for the pro photographers that they cater to, radical is scary.

[Jason Dunn]

Thanks for chiming in ptyork, always nice to have some added analysis. :)