I received my Lytro Light Field Camera back in March of this year. Unfortunately it arrived the day after I left for my vacation in Beijing, so I missed a great chance to put it through the paces in an exciting new (to me) environment. Since returning back from that vacation I’ve had a bit of time to play around with it and form some initial impressions. The announcement of this camera was met with a fair deal of skepticism from many corners of the traditional photography community. Certainly the specifications for this camera along the traditional metrics of ISO performance, megapixels, and frame rate is lacking. However I view owning this camera as potentially akin to being an initial purchaser of an original 128K Macintosh Computer back in 1984: it’s not clear yet if this is the start of a great revolution, but there is some prospect for that. Then again, there’s also the prospect that this is more like buying the much less successful Apple Lisa a couple of years earlier.
I had reduced expectations for this camera, at least as far as the conventional metrics go. I found that in good lighting conditions it was possible to get quite decent — if unexceptional — pictures with the Lytro. In my unlit studio I managed to capture this photo of my restless cat, which the Lytro auto-exposed at ISO 3200 1/60 (the camera has a fixed f/2.0 aperture). There were quite a few other blurry photos that didn’t make the cut from that session, however. One area where I was looking forward to improved performance was in the responsiveness for quick shot street-type photography. The post-exposure focusing feature of the Lytro offers the possibility of faster response times, as there is no need for slow contrast-detection autofocusing (that is used in traditional point and shoot cameras) prior to registering the image on the sensor. In this regard I got mixed results. It is true that the response time does not have the delay for focus of my Canon G12 in Auto Focus mode, however there does still seem to be more delay than I get with my DSLRs. Furthermore, it seems as if the shutter speeds chosen by the camera are slightly on the slow side for quick action photography which also makes it susceptible to showing camera shake when the shutter button is pressed. Perhaps with some more operator experience I will be able to work around this. Some form of Image Stabilization in a future version would also go a long way to making this a quick shooter.
Read on after the jump to find out why people are asking me if this camera can measure glucose levels, and what the future may hold for this technology.
I purchased the stealth grey model of the Lytro, hoping to use it as a non-conventional street camera that would not draw too much attention. I took it out to the local farmer’s market earlier today to give it a run and found that the unconventional form factor actually drew more attention than I normally get with my G12 or 5D MkII. While I was shooting a crate of turnips, one of the vendors asked me what I had there. After I explained to him that it was “just a little point and shoot camera” he confided that he thought it might have been a special monitoring device to register the amount of glucose in the vegetables that it was pointed at. I suppose it might be possible that there is such a device out there in the root vegetable management industry, but I felt so sorry for the bewildered farmer that I told him a bit more about the light field concept behind the camera. I expect he understood that explanation about as well as I took his glucose monitoring theory. My wife witnessed the conversation, and she seemed to think that the farmer was’t buying my “explanation”. Bottom line: is this camera stealthy? Not particularly when you are sticking it out in front of your face looking at the little screen trying to frame the shot, however with a bit more work on stealth techniques I am sure that it could become a better tool for street photos. The lens is right up front on the Lytro cube, however the lens focal length is closer to telephoto than it is to wide angle. I would imagine that a short black cardboard hood could be taped around the lens to make it more discrete, although I would worry about interfering with the wider light gathering required for the Lytro.
As a Version 1 hardware unit, I am impressed with the Lytro. Getting a product like this out of the research lab and into such a consumer friendly compact package is no easy task, and Lytro have made very good progress here in fairly short order. To my mind this is a promising start for the future. Where I am the most disappointed with the Lytro however, is in the post capture world. Current viewing options are constrained to the Lytro App (Mac or PC) and on the lytro.com website. It is possible to embed the “living images” from the lytro website into other websites with plugins, such as you see here. However other modes of sharing are very limited. Facebook sharing just provides a link back to the Lytro site. The end result is somewhat underwhelming: why would people bother to go visit the lytro website to click on a photo and change the focus around? Save for the initial curiosity, it is not clear what the value proposition is. To my mind, the best thing that Lytro could do here is to open up the data format to allow for independent development of other creative uses. No signs of that happening anytime soon. I have spent a bit of time looking at the files using the lfpsplitter reverse engineering tools. However it appears that Lytro is obfuscating the jpg slice image ordering behind sha1 codes, so that an automated method for reassembling them into something useful would be difficult. I understand that Lytro may not want to throw themselves completely open, however I think the clock is also ticking on this technology and surely they are looking for some more compelling uses of the technology than what they have made available so far.
My wild-ass guess as to why Lytro hasn’t opened things up is that they want to keep the Intellectual Property tightly held while they shop the technology around for a major buyer. That may not be a bad strategy. Certainly the product is going to require much more than incremental improvements to make it into the mainstream of a very competitive consumer camera market. I would imagine that sort of deep investment is difficult to build out of VC funding. Although I am fascinated by the technology and engaged by the possibilities, I am not completely sold on the Light Field Camera as a commercial venture. To me, the most compelling usage for this technology occurs at the very high end: HD and 4K video cameras. Currently “pulling focus” for video is a major chore, largely manual, that must be carefully planned and executed for each shot. The prospect of automating that process to allow changing focus during the video editing process creates an extremely compelling use case for the technology. There are a few fairly large chasms to cross between here and there however, including much higher sensor resolution, video formats, optical elements and ergonomics. Here’s hoping that Lytro can find some partners that will help them ride across the Valley of Death and into the fertile lands beyond.