The start up company Lytro plans on introducing a "light field" camera to the market this year where the focus point can be changed after the picture is taken. This type of photography captures the color, intensity and direction of individual light rays, allowing the user to refocus the picture even after it has been taken.
They also claim that they can take low light photos without a flash and 3D photos with only one lens. Pretty bold claims.
Posted by Scott Miller on June 22, 2011 at 3:38 PM
If this actually works out in a cost effective way it would be awesome. DOF stacking would become trivial, no multiple exposures needed to get perfectly sharp, in focus pixels through the while image. Awesome.
I am personally waiting for a software which needs to be fed my tripod legs co-ordinates. Whoa..my picture appears - then a few options like do you want to adjust levels, curves, dodge, burn (Say YES or NO) and you have your photo
Seeing will be believing, for sure. What was the name of that revolutionary new sensor a few years back. Oh yeah, Foveon. It was going to be in general release any day now for a good five years. Hate to be a curmudgeon, but I'm still waiting on the Foveon.
I don't think this is going to replace DSLRs soon, but I think it will be big in the point & shoot market. It sounds like it will be starting off at fairly low megapixel equivalent, which wouldn't make it practical for pro work...yet. And has Hank said, they may be making more of the technology than what it will be and it could fizzle out, but I suspect it's a glimpse of the type of technology we'll see in the future of photography.
That entry point in the market makes perfect sense to me Ryan, and makes me hopeful that it can "grow" into high end DSLRs as the technology advances with infusions of cash from the low end of the market before attacking the high end. As a car phreek friend puts it, "DeLorian's biggest mistake was to start by producing a luxury car, rather than Volkwagons. The economy car market is much larger."
I will likely be burned at the stake for heresy, but I find this disturbing. Will there come a point when analysis, synthesis, critical thought and manipulative skill are relegated to the dust bin in the face of technology that relieves us of those things? To my way of thinking, critical thought and the analysis of visual information, combined with the skills to operate a camera system and process a photograph is tied very strongly to creativity and artistic endeavor. To remove those things from the process does not appear to bode well for the continued, healthy development of the human spirit.
I'll stay with my view camera, thank you.
--P
Posted by Bill Brennan on June 23, 2011 at 12:24 AM
Preston, your points are well made. However, the die seems to be cast; sad as it may be. Quants and their algorithms are penetrating virtually all decision making. Just as there is a small market today for film and medium and large format cameras, so will go the market for digital slrs and related equipment; a slow loss of market share.
Those are illuminating insights on the creative process meeting technology, coming from folks that know what they're talking about.
I agree to the points when applied to artistic photography, but it could shine for other genera. I can see where our industrial, event, wedding, sports and even wildlife photography could profit, to name a few- basically any situation where opportunities are fleeting and there's no "Take Two" or waiting for better light. If you have to get the shot and no excuses allowed, it holds a lot of promise.
This reminds me of Douglas Adams' novel Mostly Harmless, where Arthur Dent has a daughter who was raised on flying through space and advanced technology. She gets a hold of his wind-up wristwatch, which confuses her, and has one main question :
"Why's it all in hardware?"
Will there come a point when analysis, synthesis, critical thought and manipulative skill are relegated to the dust bin in the face of technology that relieves us of those things?
I'd say that nothing about current technology is relieving you of any of those things. They are simply shifting where and when and how you apply them. People were already only applying them when they chose to. P&S cameras have long ago gotten rid of any real thought being necessary to take photographs.
I totally agree with Preston, can't we do anything by ourselves anymore. What's going to happen when all this technology stops working one day?? We'll probably stand there with this blank look on our faces, which reminds me of the late 60"s , and this is coming from someone who considered himself a Technocrat at one time.
I saw thousands of similar "high technology" announcements as part of my job for 30 years. The vast majority never even made it to the production prototype stage because an announcement like this is all about trying to get someone to provide the millions of dollars that are needed to build and test that production prototype.
The image that Cam posted in the NPN forum looks good, but I doubt that you'd get anything like that quality if the two subjects were separated by a significant distance (say 10 feet, to pick a random number). In that case, I think the problem will be the limited amount of information (small number of pixels) that the distant object occupies. It also seems to me that we already know how to "deblur" an image on the computer provided that we know the transfer function of the lens, so I wonder what is this doing that's so different?
Posted by Joe Abounader on June 23, 2011 at 10:05 PM
Preston, I'm in your camp. I still prefer to put most of the effort in the field, with my camera, and use technology to develop the photo, not make it. Granted the lines are already blurred. There are however lines that I hope I will never cross; this would be one of them.
Mark - you cannot deblur an image. You can, however, blur an image to simulate a given lens/aperture. But, blurring is not a reversible process.
This technology is different in that it captures information that is ordinarily discarded - it captures the entire light field, rather than just the light intensity at a single plane. It essentially allows them to do backwards ray-tracing to reconstruct the 3D world (but from only one perspective), and then re-photograph that reconstruction using different simulated lens parameters. There is no free lunch, so you pay for the additional depth information with planar resolution - ie. the light that used to hit one pixel now needs to be spread out over many, many more. But, as pixel densities increase, this will become less and less of a drawback.
The images on their website are actually pretty impressive in terms of the capability, but you can find some examples of the limitations as well: Lytro picture gallery
Posted by Paul DeCesare on June 24, 2011 at 11:32 PM
I find this subject interesting on this site, considering we strive for sharpness from 2 inches to 2 miles, lol. After thinking on it for a while, I guess I just don't want that much automation. There's a certain sense of accomplishment when you nail a shot regarding focus, composition and exposure. AF was amazing in 1985 when it showed up, now we turn it off so we have control. It's all good regarding these technologies, used when and where they should be. If I was a wedding photographer I'd be overjoyed at Lytro's stuff, no more missing that once-in-a-lifetime look on the bride's face due to the 85mm/1.2 focusing too slow!
Posted by Basil Greber on June 25, 2011 at 6:16 AM
I'm not sure whether I get Hank's point. Foveon sensors have been in Sigma's DSLRs for quite a while now. They're out there, people buy them, and they have some advantages over conventional Bayer filtered sensors...
I'm not sure whether I get Hank's point. Foveon sensors have been in Sigma's DSLRs for quite a while now. They're out there, people buy them, and they have some advantages over conventional Bayer filtered sensors...
My point Basil is more about the large hype preceding the eventual release of the Foveon than the sensor itself. It's not that it doesn't work, rather that it was hyped long before release just like this latest widget, as revolutionary and sure to change photography as we knew it. All the DSLR's would be incorporating Foveon, etc, etc. Foveon took several extra years beyond promises to actually appear in a camera after it was announced, and in fact no other manufacturer has picked it up that I've heard of. There's been no "evolutionary" move down that particular path either, and I'm yet to see a photo that was a standout due to the technology or to hear a photographer say anything like "Thank goodness I was using Foveon."
Foveon looks to me like a very short "side branch" in the evolution of photography, and almost certainly a branch that will grow no further. I'm wondering if the same fate awaits Lytro.
Posted by Basil Greber on June 26, 2011 at 2:03 PM
Thanks for the clarification, Hank. Makes a lot of sense.
No one should get me wrong on my opinion of the potential of the technology, should it live up to promises already being made. If it works as billed I can see direct application in some of our work. Should it become available in a DSLR body that meets our needs, I'll almost certainly become a loyal customer.
But until that happens, I'm not going to get my hopes up about function or availability.
Would I use it for art photography? If it was in my bag and the opportunity arose, certainly. I doubt anyone would refuse to take the photo because they had the "wrong" camera in their bag.
Discussion
24 repliesThey also claim that they can take low light photos without a flash and 3D photos with only one lens. Pretty bold claims.
A link to the story: http://www.pcworld.com/article/230867/camera_startup_lytro_promises_to_revolutionize_photography.html
Sandy
. It hopes to have a point-and-shoot model on sale later this year. The device will be “reasonably priced,” but Lytro didn’t offer further details.
“It will be a competitively priced consumer product that fits in your pocket,” Ng said.
Hope it has at least 12-16 MP and can capture RAW. Too much to ask?
Sandy
http://www.fredmiranda.com/forum/topic/1007950
Sandy
I'll stay with my view camera, thank you.
--P
Bill Brennan
I agree to the points when applied to artistic photography, but it could shine for other genera. I can see where our industrial, event, wedding, sports and even wildlife photography could profit, to name a few- basically any situation where opportunities are fleeting and there's no "Take Two" or waiting for better light. If you have to get the shot and no excuses allowed, it holds a lot of promise.
"Why's it all in hardware?"
I'd say that nothing about current technology is relieving you of any of those things. They are simply shifting where and when and how you apply them. People were already only applying them when they chose to. P&S cameras have long ago gotten rid of any real thought being necessary to take photographs.
The image that Cam posted in the NPN forum looks good, but I doubt that you'd get anything like that quality if the two subjects were separated by a significant distance (say 10 feet, to pick a random number). In that case, I think the problem will be the limited amount of information (small number of pixels) that the distant object occupies. It also seems to me that we already know how to "deblur" an image on the computer provided that we know the transfer function of the lens, so I wonder what is this doing that's so different?
This technology is different in that it captures information that is ordinarily discarded - it captures the entire light field, rather than just the light intensity at a single plane. It essentially allows them to do backwards ray-tracing to reconstruct the 3D world (but from only one perspective), and then re-photograph that reconstruction using different simulated lens parameters. There is no free lunch, so you pay for the additional depth information with planar resolution - ie. the light that used to hit one pixel now needs to be spread out over many, many more. But, as pixel densities increase, this will become less and less of a drawback.
The images on their website are actually pretty impressive in terms of the capability, but you can find some examples of the limitations as well:
Lytro picture gallery
My point Basil is more about the large hype preceding the eventual release of the Foveon than the sensor itself. It's not that it doesn't work, rather that it was hyped long before release just like this latest widget, as revolutionary and sure to change photography as we knew it. All the DSLR's would be incorporating Foveon, etc, etc. Foveon took several extra years beyond promises to actually appear in a camera after it was announced, and in fact no other manufacturer has picked it up that I've heard of. There's been no "evolutionary" move down that particular path either, and I'm yet to see a photo that was a standout due to the technology or to hear a photographer say anything like "Thank goodness I was using Foveon."
Foveon looks to me like a very short "side branch" in the evolution of photography, and almost certainly a branch that will grow no further. I'm wondering if the same fate awaits Lytro.
No one should get me wrong on my opinion of the potential of the technology, should it live up to promises already being made. If it works as billed I can see direct application in some of our work. Should it become available in a DSLR body that meets our needs, I'll almost certainly become a loyal customer.
But until that happens, I'm not going to get my hopes up about function or availability.
Would I use it for art photography? If it was in my bag and the opportunity arose, certainly. I doubt anyone would refuse to take the photo because they had the "wrong" camera in their bag.
US Patent Office, 1902.