Shared publicly  - 
 
My first photowalk with the Lytro lightfield camera

This weekend we'll be in Yosemite with Google+ photographers, including +Thomas Hawk +Trey Ratcliff +Karen Hutton +Scott Jarvie +Peter Adams +shirley lo and +Kimberly Shoemaker. All trying to make a better image than the ones that come up in searches for Yosemite: https://plus.google.com/s/Yosemite

So, to prepare, I went on a photowalk around Half Moon Bay with a new kind of camera: +Lytro's lightfield camera. Lytro's director of photography, +Eric Cheng, gave me a late Christmas present: he let me be one of the first people in the world to actually use one in the wild. Yesterday we walked around and I made 278 images. He shot a bunch too (he's one of the world's best underwater photographers, by the way, and has videos of sharks eating GoPro cameras, crazy!). Unfortunately Eric didn't let me share the actual images online, since this is still a pre-production unit and doesn't have the final software or viewer yet, but we did shoot a video where you can see some of the results.

I have already purchased a Lytro with my own money, so you know I'm interested in this new camera, that lets you do things like refocus images after you shoot (it does more, too, which we discuss in the video).

So, how was it?

Both disappointing and enthralling.

But first, this is NOT a review. It's just an early look at a product that hasn't yet shipped (they expect to ship them sometime in Q1, 2012, so by April 1, although first units might start shipping in February). The software isn't done, and Eric showed me a few things that they are working on for the future.

You'll have to wait for an official "review" of the final camera.

So, why was it disappointing?

Well, if you just want the ultimately sharpest photo, this isn't a camera for you (it won't do 22 megapixel photos like my Canon 5D MKII will, and the images are generally good enough for on-screen use but if you want to blow them up to wall sized images, this isn't a camera for you).
If you like having a huge choice of lenses, this isn't a camera for you.
If you want to shoot action sports, this isn't a camera for you.
If you want to see through the viewfinder to choose your own focus point, this isn't a camera for you.
If you want the best low-light performance, then this isn't a camera for you.

But why is it enthralling?

It let me see the world in a new way. I no longer needed to worry about focus. In fact, I quickly learned that there's a kind of photo that only works on the Lytro: one where you can get very close to the subject and just shoot, without any worry about where the focus is.

Plus, coming sometime after the camera ships you can turn each image into a 3D image. I saw some examples from Eric's computer on my 65-inch Vizio 3D TV and they rocked.

Some other reactions.

1. Shooting is actually pretty comfortable and fun. In the video you'll see Eric shooting with it.
2. The shutter reacts pretty quickly. I was able to capture some shots of golfers in mid swing. That said, top shutter speed is 250th of a second, so this won't freeze most sports action. Water that I shot out of a fountain was slightly blurred because of the slow shutter.
3. Exposure was usually pretty good, although on some subjects, highlights were overblown. Eric says that they are still tweaking the settings in the camera, so these will probably improve.
4. The viewfinder was frustrating to use in bright sunlight. In fact, most of the time I just shot without seeing the image. That isn't as big a deal as it might seem, though, because you don't need to focus, just need to properly compose the image. Eric says that they are working on making the viewfinder brighter.
5. In low light images got a big grainy for my tastes, but still worked.
6. To get the "refocusable effect" you need to pick images where the camera is extremely close to one subject while another subject is in the distant background. This takes a little bit of playing around to optimize for, but I got some good examples, including one where I stuck the camera four inches away from a window frame and shot outside.
7. The camera gets a lot of reactions. At one point the bartender at the Ritz grabbed ours and said "I read about this in Wired" and started shooting with it. The fact that he could pick it up and figure out the controls quickly tells me it is well designed.
8. There are improvements coming that I can't talk about.
9. Some images have light-field artifacts. This happens when it can't build the 3D model properly that it relies on, like when there's motion blur. These aren't going to be noticed by most people who view your images on Facebook but we were blowing the images up on my 65-inch TV.

Is this camera worth buying?

For me and other early adopters who want to own a piece of the future, absolutely 100% yes.
For my wife? She'll probably keep using her iPhone's cameras.

I can't wait to get mine for real. The technology behind this is mind blowing.
263
195
Hans Watson's profile photoBlair Evan Ball's profile photoMILOS JANATA's profile photoPeter Harris's profile photo
74 comments
 
I saw you holding it over the pond. I'm glad you have a firm grip. 
 
More power to all the budding Ansels.........it is all about the timing and the planning....or 30,000 strobes to playboy light el capitan....
 
You can see the images starting at about six minutes into the video, by the way.
 
http://www.lytro.com/ pre-orders not working for the UK, maybe you guys can in the US though, great report Scoble, just had blogs/articles so far.
 
I ordered mine the first day you posted about it on G+ (a few months ago). I'm looking forward to trying it out.
 
Sorry for the wall of text, I'm just really curious about a few of the technical things, since I won't be able to play with one of these things for quite some time.
* Just how bad is the low-light performance? I mean, is it at least usable in a normal room with one or two lamps on? (not that the camera is meant for that kind of use, but just as a sort of baseline for comparison)

* About the snapping pace: It takes pictures pretty quickly, I get that, but what is the shutter lag like? And how long does it take to "process" the image (that is, how fast can I snap shot after shot, rapid fire?)

* A thought on the viewfinder problem: Why not just go old-school with it and have just a pop-up aiming crosshair? Just so you can line up the shot and use that, kind of like viewfinders on disposable cameras, except I'm picturing it in a housing much like the pop-up flashes in a lot of cameras. It's not high tech, but as you say: you just need to get your framing about right.

* Exactly how close can you get to something with the Lytro? I mean, four inches is impressively close, but does it even have a minimum "focusing" distance?

* Last question: what was color reproduction like? I can guess that since you haven't mentioned it here, it can't be extraordinarily bad, but I'm a little interested to see where it ends up, in general.

Again, sorry about the wall of text, but new kinds of imaging technology like this really excite me.
 
I remember seeing this technology a while ago i tried to talk to people about it but their eyes rolled over as usual. I think it defiantly has a big future. Would be possible for someone to create a format for the image so the audience can play with the focus or allow for animation?
 
+Kimberly Hayworth, thanks for the +ing! +Robert Scoble, thanks for being the canary again for all of us :D No, wait, that's not the right metaphor. The dog? Hmm, no, dangit... er, the fox? :)
 
Been keeping up on this for the last year or so, its been really interesting. I wont be buying one anytime soon as it doesn't sound like it will aid my work, maybe in time when more works been done on them and they are more usable for sports and low light.

saying that amazing step forward for photography and generally in technology.
 
+Ivan Kowalenko it's very usable, yes. I shot a bunch in low light and it was fine with most of them. In some, though, the grain is apparent.

Colors were good to very good. I need to go out with one with a Canon 5D MKII to see the difference in how they approach colors, though.

I totally could see using a gunsight-like device to line it up.
 
+Euro Maestro I made Eric nervous when doing that. He joked that I'd be the first one to drop it into water.
 
(I apologize if these are answered in the video but I couldn't get it to play all the way through) What is the basic pixel dimensions of a picture taken with the Lytro? I know they use a unique format to allow for the refocusing but will it output to raw after the refocus? Are you retaining all that color info?
 
I'm so interested in this technology. Thanks for your non-review or whatever you're calling it ;)
 
Is the image translation algorithm ("codec") open source?
 
+Robert Scoble: Thanks for the answers. Sounds like they have a product solid enough to stand up to some criticism without being completely ruined. I mean, let's face it, almost all First Generation technology has some bugs to work out, but that's part of the fun of being an early adopter: the excitement of testing the limits of something completely new.
 
+Hans Watson it's 1080x1080. Yes, the file format keeps everything in raw, so it can be improved on later (and even output to 3D). Yes, everything is kept, even after a refocus.
 
+Ivan Kowalenko it depends on how you look at it when you review it. I look at things like these from this point of view: "does it let me do something your camera doesn't?" On that point of view this will be a huge success.
 
I'm confused -- I skipped through the video, but didn't see any pictures from your photowalk. Where are they hiding?
 
Exciting new technology, still has a way to go, but the possibilities are endless. Is it possible to have the whole image in focus, not just one area where you click?
 
+Blair Evan Ball yes, that's possible. They aren't sure if they will ship that capability in the viewer, though.
 
+Robert Scoble: True. Not everyone will like it. But being that the things it does well, it does excellently; and the things it doesn't do well, it doesn't completely fail at is rather reassuring. Hopefully the Lytro folk will be able to generate enough money to continue to develop the technology, because it sounds like most of the camera's shortcomings can be eventually improved, given enough time and money for R&D.
 
excellent commentary and coverage. Your point is excellent. In some production work we call the "refocusing" to be foreground establishment ( film-makers kind of term ) or subject establishment, so often the value of any image is in the choice of DOF for subject matter. It appears the Lytro goes well for stable lighting so such artistic stuff is available to afterthought. I definitely prefer the tone and treatment of way you do these ( and how you cram softboxes in ... i have the same stuff ), really its the best so thanks for that.
 
I think it will be interesting to capture some of the HDR guys impressions on these, and the possibility of mixing both exposure and focus across the 3 (or more) shots used in the creation of a super-amazing HDR photo.
 
There are some issues that I read about the Lytro

1. Low resolution images. Maybe intended only for web viewing.
2. Software limitation. Can you select focus point and then save as PNG or JPEG? That way, you have practically an unlimited number of images per capture RAW image. OR does it require a plug-in to be viewed? I remember Kodak's FlashPix format before.

Still on the fence regarding this gadget but sure would love to see it on an iPhone. LOL.
 
I suppose everything in one of these images could also be in focus simultaneously.
 
Very informative preview - good to hear it's not too hard to shoot with. The form factor seemed kind of odd. One other thing - you mentioned motion blur from moving objects, but what about camera shake, is that an issue? Is it necessary to set it on a surface/tripod in mid- or low- light?
 
That's a very good point +Rob Crawford in theory the technology should be able to dynamically adjust the depth of field. I wonder if their little web player supports that though
 
Robert, glad you had some fun with the Lytro. Ordered mine first day also...looking forward to it. thanks
 
Looks interesting and very simple clean design. If I had the money I would get that.
 
Ha, define art.

Although, I can imagine a totally automated image capture process where a device just captures everything all the time, and you have software that automatically selects the best ones based on decision parameters set by the user. For a lot of people, that's really all they want.
 
+Robert Scoble But this is just the start of the journey so the disappointments will be eroded over time, shutter speed will get faster, etc, etc,. Just imagine a video based on this technology - no more blurry security footage of the holdup - run the tape and select the mug to focus on during playback. Sports photography, it will come with longer optic choices and enhanced sharpening algorithms and/or yet smaller micro-lenses in the sensing array. After all, this is just the first generation device - look at the vast array of film/digital cameras and formats that exist - Lytro will follow them into specialist cameras for specialist applications and the images will get bigger and better.
 
Interesting toy. Success depends on the fact if users feel inspired or limited.
Video feature must come soon...
 
Winter is my favorite time to visit Yosemite, providing there is an abundance of snow on the ground.
 
I like it and if only I own it once!!!
 
I read about this in Popular science. This shoulc be put inside Cellphones.
 
This should be inside a phone ..somehow..
 
I hope I'm in the Feb shipment! That's what I was told when I pre-ordered. Looking forward to having fun with it as a late birthday present. :)
 
Can you process the picture to be focused everywhere ?
 
Hey, +Itai Frenkel, about 12 minutes into the video Eric says that "focus everywhere" is coming (also 3D). The good thing is that these are just software updates!
 
Not sure if people notice but 3D movies have an annoying aspect of not being able to focus the way you would in a real life. The movie director has decided for the viewers what should be focused which is typically the foregound. But imagine using two of these to take a 3D image and monitoring the viewers eye lense (probably a extremely low power laser) and some software to control the focusing will allow at least for the user to focus just like a normal real life scene.
 
+Vilas Rokade Had a long conversation with a friend about this camera and came to the conclusion that this technology is the forerunner to bionic eyes of the future.

+Robert Scoble Would you technically call them "photos"? They seem to act more like reverse holograms. Awesome technology nonetheless.

(Edited as I was able to find answers to my posed questions.)
 
+Mike Greenberg Mike, the resolution is definitely not fixed at 1080 x 1080. We just ask our light field engine to output an image at that size. Output resolution in 2D is not very straightforward, and is a difficult concept to communicate. It's better to say that it is an "11-megaray light field picture." Over time, people will understand what that means as 2D projections.

You're right that the capture is effectively pseudo-holographic. A light field picture is not a hologram, per say, but we can print holograms from them.
 
Sounds like a fun day. Nice to see and hear about expansion of photography using lightfield technology. Exciting!
 
Theoretically the only realistic and feasable way a lazer could accurately track the direction a pupil would be through a contact lense or Glasses with a dlp chip and camera that can that projects the focal point on a display that can read precise IR reading. Doing multiple views would take the same technology that the PS3 3d display uses.
 
The Simulview technology and extremely tight video syncro.
 
can we expect this to come in the form of kodak's viewmaster type shape eventually
 
+Mike Greenberg Damn right about the bionic eyes.

Processing 3D images using the regular dual camera setup was an exciting part of computer vision in the 2000's. As always, we started with what we knew from evolution. But evolution is fairly shortsighted (pardon the pun). This kind of cheaper imaging technology however should completely change the way computer vision and cognition evolves. With this it should not be that hard to recognize objects that are far and ones that are near, define their outlines etc. Using the dual camera setup to resolve foreground objects, define their outline etc, is computationally intensive. If this technology fits in a battery powered device you know that it is computationally far simpler now.

Can't wait for the car to drive itself :-)
 
Why does he have 3 apples behind him?
 
That's my home studio and I have three Apple monitors.
 
Ok... Now... that doesn't make me feel too dandy staring at MY 3 PC monitors, an LG, Samsung and HP - but that's life. One day when I grow up I will be just you bro :}
 
this post and its comments is probably the best place to get a real overview of the Lytro lightfield camera! I hope it will be soon avalaible in Europe.
Add a comment...