fredmiranda.com
Login

Moderated by: Fred Miranda
  New fredmiranda.com Mobile Site
  New Feature: SMS Notification alert
  New Feature: Buy & Sell Watchlist
  

FM Forums | Leica & Alternative Gear | Join Upload & Sell

       2       3       end
  

Archive 2011 · Lytro

  
 
crraynes
Offline

Upload & Sell: Off
p.1 #1 · Lytro


Has anyone seen this yet? I am curious as to what you think of it. Worth buyng stock in, or is it one of those "promises" that we will never see come to fruition?

http://lytro.com/about/index.html



May 10, 2011 at 01:20 PM
AhamB
Offline
• • • • •
Upload & Sell: Off
p.1 #2 · Lytro


It reminds me a bit of what that Canon future concept camera that was presented a while ago was supposed to be able to do. Taking the creativity out of the shooting and choosing focus/DOF/composition in post. Sounds great...


May 10, 2011 at 01:30 PM
mpmendenhall
Offline
• • • •
Upload & Sell: Off
p.1 #3 · Lytro


Given that the site contains virtually no information, other than what appears to be a marketer-babble translation of existing synthetic focus research, I wouldn't be in a hurry to open my wallet. At least wait until you've finished any current transactions you have with friendly Nigerian princes before jumping on this next "opportunity."


May 10, 2011 at 01:37 PM
jotdeh
Offline
• • •
Upload & Sell: Off
p.1 #4 · Lytro


Adobe presented a camera like this back in 2007 at siggraph and even then it wasn't new. Don't put too much hope in this. It's essentially an array of lenses either over one sensor or multiple sensors (-> multiple cameras). In either case the "virtual" aperture is not continuous but full of holes, and out of focus areas may suffer from artefacts like those on mirror lenses, i.e. nasty bokeh.


May 10, 2011 at 03:13 PM
Lotusm50
Offline
• • • • •
Upload & Sell: Off
p.1 #5 · Lytro


mpmendenhall wrote:
Given that the site contains virtually no information, other than what appears to be a marketer-babble translation of existing synthetic focus research, I wouldn't be in a hurry to open my wallet.



Yes, the site has surprisingly little information. Nothing on the people involved, nothing on the technology, nothing on patents and other IP positions, nothing on prototypes, nothing on where they actually stand in the development of the technology/product, no real hint on where this potential tech/product would be positioned Given that this site is essentially investor oriented (at the very least, by default), there is nothing here that would get an investor interested.




May 10, 2011 at 03:37 PM
freaklikeme
Offline
• • • • • •
Upload & Sell: Off
p.1 #6 · Lytro


Those sample images aren't exactly making me do backflips (not that this is possible, but it's not inspiring the desire to try). It seems like you've got a choice of getting a subject mostly in focus or picking an abstract position in the photo and getting nothing truly in focus.


May 10, 2011 at 06:23 PM
LightShow
Offline
• • • • •
Upload & Sell: Off
p.1 #7 · Lytro


An interview with the CEO of Lytro, and a camera to be released this year.
http://www.1001noisycameras.com/2011/06/rd-light-field-stealthy-start-up-lytro-gets-into-the-spotlight.html

http://www.lytro.com/picture_gallery



Jun 22, 2011 at 12:38 AM
mpmendenhall
Offline
• • • •
Upload & Sell: Off
p.1 #8 · Lytro


OK, so this confirms it's a microlens array based approach (i.e. no shocking new developments). So, what does this mean? Well, "there's no such thing as a free lunch." Grouping pixels together behind a microlens array in order to extract direction/depth information means trading off 2D spatial resolution in exchange for the additional 3D planes. Suppose, for example, that the camera uses a 5x5 pixel array behind each microlens; then a modern 25 megapixel sensor would deliver a whopping 1 megapixel image when "focused" to a particular position. The tiny, marginally sharp, web-sized examples shown may well be all this technology can deliver. I suspect this is why the promotional material indicates that they are going after the "consumer" easy point-and-shoot market, because the camera is only capable of producing small, gimmicky web-sized photos.

Furthermore, the microlens array can only be matched to one lens aperture; if a faster lens is used, the additional light is "thrown away," while a slower lens will leave sensor pixels unused (and reduce the "depth" information available for reconstructing the image), so interchangeable lenses will be rather limited. For professional through even minimally skilled amateur work, needing to post-adjust focus isn't a big concern (most folks can usually get focus close enough to right on the first try), and certainly not worth taking a HUGE hit in resolution for. High quality 3D work will still be done with multiple-camera setups.

I think the technology is really cool from a nerdy computational/data reconstruction aspect, but I don't see it either revolutionizing the general world of photography or providing particularly useful cameras for intermediate amateur to professional photographers. Until gigapixel camera sensors are the norm (providing enough "extra" photosites that one can easily afford to "throw away" a lot of 2D spatial resolution in exchange for other measurements), this type of technology will be limited to gimmicky, web-sized snapshots (which may indeed be a big market for selling cameras, but doesn't do much for me).



Jun 22, 2011 at 01:21 AM
LightShow
Offline
• • • • •
Upload & Sell: Off
p.1 #9 · Lytro


Suppose, for example, that the camera uses a 5x5 pixel array behind each microlens; then a modern 25 megapixel sensor would deliver a whopping 1 megapixel image
120mp sensor
http://www.dpreview.com/news/1008/10082410canon120mpsensor.asp
=4.8mp image
Would need 250mp sensor to reach 10mp image, assuming you would need 5x5 array.

I agree that until we get 200mp+ sensors this will be gimmicky, but I could see them in cellphones with say a 3x3 array,
web cams that can focus on faces with no moving parts.
Which could be very popular.






Jun 22, 2011 at 03:49 AM
RalphJ
Offline
• • •
Upload & Sell: Off
p.1 #10 · Lytro


http://www.nytimes.com/2011/06/22/technology/22camera.html

Be sure to click on the "[image] available here" in the caption under the flower photo to see a "refocusing" example.



Jun 22, 2011 at 07:39 AM
mpmendenhall
Offline
• • • •
Upload & Sell: Off
p.1 #11 · Lytro


LightShow wrote:
I agree that until we get 200mp+ sensors this will be gimmicky, but I could see them in cellphones with say a 3x3 array,
web cams that can focus on faces with no moving parts.
Which could be very popular.


However, this technology doesn't work well with the tiny sensor formats used in phones/web cams, for a couple of reasons. First, focusing isn't even much of a problem on tiny sensors; the few-mm focal length lens used in these applications already has a huge depth of field, and shows very little difference whether in or out of focus. Second, the tiny sensors are already pushing the limits on pixel pitch size imposed by optical wavelengths --- you can't get any extra information (2D or 3D) once you go smaller than the wavelength of light (sub-micron pixel sizes). Only by making a 35mm or larger sensor using cellphone-sensor pixel sizes can you get to the 200Mp+ sizes needed to do much with this technology.

On the Lytro site, they actually have Ren Ng's (Lytro founder) graduate thesis available for download, which explains the real science/development behind the camera instead of the vacuous marketing/investor fluff that fills the website and press (advertising/PR) pieces. The thesis starts by suggesting this as an application for 250+ Megapixel 35mm-format (or larger) sensors. Throughout the paper, a target output of ~2.1Mp (as a theoretically achievable goal) is suggested. The prototype constructed used a 16Mp medium-format Kodak back on a Contax 645 body, with a 296x296 (=0.088Mp) microlens array (about 13x13 sensors behind each lens). Also interesting is the need for ~f2 diffraction-limited optics (far better than currently commercially available 35mm lenses) to make this work when scaled up to "full" resolutions, although the directional information allows for some corrections for lens aberrations.

This technology might make a decent point-and-shoot in two or three decades (once super-high-resolution full-frame sensors are cheap and assuming significant improvements in lens manufacturing). At this point, it looks mainly like an attempt to collect a lot of investor money (and possibly secure a big patent portfolio for licensing fees), in order to produce a mid-priced gimmick aimed at non-photographers.



Jun 22, 2011 at 08:57 AM
Arun Gupta
Offline
• •
Upload & Sell: Off
p.1 #12 · Lytro


June 22, 2011 artice in the New York Times
http://www.nytimes.com/2011/06/22/technology/22camera.html

Key excerpts:
Quote:

With an innovative camera due out later this year from a company called Lytro, photographers will have one less excuse for having missed that perfect shot.
....
Mr. Ng explained the concept in 2006 in his Ph.D. thesis at Stanford University, which won the worldwide competition for the best doctoral dissertation in computer science that year from the Association for Computing Machinery. Since then Mr. Ng has been trying to translate the idea into a product that can be brought to market — and building a team of people to do it.

The Lytro camera captures far more light data, from many angles, than is possible with a conventional camera. It accomplishes that with a special sensor called a microlens array, which puts the equivalent of many lenses into a small space. “That is the heart of the breakthrough,” said Pat Hanrahan, a Stanford professor, who was Mr. Ng’s thesis adviser but is not involved in Lytro.

But the wealth of raw light data comes to life only with sophisticated software that lets a viewer switch points of focus. This allows still photographs to be explored as never before. “They become interactive, living pictures,” Mr. Ng said. He thinks a popular use may be families and friends roaming through different perspectives on pictures of, say, vacations and parties posted on Facebook (Lytro will have a Facebook app).

For a photographer, whether amateur or professional, the Lytro technology means that the headaches of focusing a shot go away. Richard Koci Hernandez, a photojournalist, said that when he tried out a prototype earlier this year, he immediately recognized the potential impact.

“You just concentrate on the image and composition, but there’s no need to worry about focus anymore,” Mr. Hernandez said. “That’s something you do later.”

“That was the aha! moment for me,” said Mr. Hernandez, an assistant professor of new media at the graduate school of journalism at the University of California, Berkeley. “This is game-changing.”

Mr. Hernandez, who is not affiliated with Lytro, was one of several photographers who tested prototypes. His model, he said, was sheathed in a black plastic shell, so he did not see its design. But he said it was the size of a standard point-and-shoot camera. The picture resolution, he added, was indistinguishable from that of his other point-and-shoots, a Canon and a Nikon.

Eliminating any loss of resolution in a camera like Lytro’s, which is capturing light data from many angles, is a real advance, said Shree Nayar, a professor at Columbia University and an expert in computer vision. Mr. Nayar is familiar with Mr. Ng’s work, but he said he had not seen anything Lytro has done in more than a year.
....
Among its other advantages, the new camera is much faster than conventional ones because there is no “shutter lag” — waiting for the autofocus device to work and the shot to be taken. ....

Lytro cameras can also capture plenty of data for 3-D images, which can be viewed on a computer screen with 3-D glasses.

Lytro is not saying what the price of its first camera will be, but insists it will be for the consumer market, which suggests a price of a few hundred dollars. ....









Jun 22, 2011 at 09:47 AM
Ataboy
Offline
• • •
Upload & Sell: Off
p.1 #13 · Lytro


If this goes in the right direction, in relatively short timeframe we might have a camera that can take *any* lens! Imagine the possibilities:

- how compact and light lenses could become if there if no need to have a focusing helicoid, AF motor and ever move lens elements relatively to each other? Put the aperture diaphragm into the camera and the lens becomes just a simple tube with few pieces of glass inside.
- no need to worry about register distances anymore, perhaps you could even use those uber-fast X-ray lenses for normal shooting
- use manual focusing lenses in "autofocus" mode (Contax AX Digital, here you come!)
- video capabilities: perfect focus every time (I've ruined so many videos on my 5D II because of misfocusing...)

Cool, huh?



Jun 22, 2011 at 11:08 AM
Pavel
Offline
• • • • •
Upload & Sell: Off
p.1 #14 · Lytro


We knew that our digital camera toys were always outdated soon - but this budding technology which is more profound than the change from manual focus to autofocus ever was makes me feel, more than ever, the silliness of getting caught up in any marginal upgrades. Ahhh ... the great breakthroughs we have all gotten tied up in over the years as if they were religion. 1.5 megapixels, two megapixels, FOUR! next year ... 26, 28 ... perhaps even ...gasp .... 29! No wonder my pictures have stunk - I only have 12.

The real improvements over the years have been dry plate, film, digital. In-between has I believe just been the much beloved hype loved so much by marketers. Video is the current next step for photography and this development, if it pans out the one after that.

I guess it's good. Something immature to be refined over generations of iterations to keep us in the spin cycle. Gotta love it and keep our fingers crossed. Just in time. Dynamic range and megapixels are getting seriously boring - don't you think?



Jun 22, 2011 at 11:19 AM
CGrindahl
Offline
• • • • • •
Upload & Sell: Off
p.1 #15 · Lytro


Here's an interview with the developer of this technology. I love the photos on the company's website... very engaging.

http://photorumors.com/2011/06/22/startup-company-lytro-gets-funding-promises-to-change-photography/#more-11132



Jun 22, 2011 at 11:21 AM
CGrindahl
Offline
• • • • • •
Upload & Sell: Off
p.1 #16 · Lytro


mpmendenhall wrote:
...At this point, it looks mainly like an attempt to collect a lot of investor money (and possibly secure a big patent portfolio for licensing fees), in order to produce a mid-priced gimmick aimed at non-photographers.


You mean the hundreds of millions of folks taking photos with their cell phones who have no interest in Leica, Zeiss or much else discussed on this website? Purists may not care for this technology, but that hardly is reason to dismiss it. This may very well be the direction photography will take for the future and for one simple reason... It IS simple and engaging. The developer in an interview noted that 60 BILLION photos were uploaded last year on Facebook. Obviously, for most of these people, having a photo that can be manipulated would be more appealing than one that can't. In the same way professional photographers are finding it ever harder to make money selling their photos because millions of people are taking billions of photos, albeit of a lower quality, the market for quality is fast disappearing. Call it Gresham's Law if you will. Bad money drives good money out of the market.

By the end of this year we'll find out whether this is a great strategy for fleecing investors, or whether this guy and his company will deliver a game changing camera. Inquiring minds want to know...



Jun 22, 2011 at 11:30 AM
joeinla
Offline
• •
Upload & Sell: Off
p.1 #17 · Lytro


This thing is going to be HUGE. Yes, it's a "facebook" camera. Oh yeah, 1/5 of the whole world is on facebook....

I think the market for "serious" cameras is infinitely smaller :P

Each of us will almost certainly buy one for our families. And we'll use them. But I'm pretty sure we're not selling our Canikons for this... At least not for 10 years or so :P



Jun 22, 2011 at 12:11 PM
jotdeh
Offline
• • •
Upload & Sell: Off
p.1 #18 · Lytro


Yes I think the snapshot market is a big one, and with current P&S cameras being largely equal (most comparisons in the purchasing process still coming down to megapixels, zoom and body colour), this will be a very attractive feature. Also, as they Arun quoted, it can be very small (essentially very short focal length lenses clustered closesly together) and STILL give shallow depth of field. That might be an interesting feature even to purists.

I'd give this the edge over 3D cameras that require special displays or glasses.

I imagine a zoom would be implemented digitally.

I don't think making this work with interchangable lenses / lens arrays is feasible nor necessary at this point to generate sufficient interest.



Jun 22, 2011 at 12:57 PM
abhijeeth
Offline
• •
Upload & Sell: Off
p.1 #19 · Lytro


Thom Hogan has an interesting post on his site today about this. Although he talks more about the business side of things, he does have a link to more technical info: the PhD dissertation written by the company founder.


Jun 22, 2011 at 02:29 PM
curious80
Offline
• • • •
Upload & Sell: Off
p.1 #20 · Lytro


mpmendenhall wrote:
However, this technology doesn't work well with the tiny sensor formats used in phones/web cams, for a couple of reasons. First, focusing isn't even much of a problem on tiny sensors; the few-mm focal length lens used in these applications already has a huge depth of field, and shows very little difference whether in or out of focus. Second, the tiny sensors are already pushing the limits on pixel pitch size imposed by optical wavelengths --- you can't get any extra information (2D or 3D) once you go smaller than the wavelength of light (sub-micron pixel sizes). Only by making
...Show more


What you are saying is true. However that work was done 5-6 years ago and we dont know what additional developments and refinements have been made during this time. The NY article does suggest that Lyto guys found a way to reduce the resolution loss - lets see if thats just a marketing ploy or a reality. In any case I believe concepts like this will get explored more and more, and computational photography in general will play an increasingly larger role in years to come.



Jun 22, 2011 at 07:25 PM
       2       3       end




FM Forums | Leica & Alternative Gear | Join Upload & Sell

       2       3       end
    
 

Welcome back
Log in to your account