The day will come, you will forget about what the word “autfocus” means. And looking at the newly announced Lytro Illum camera (https://www.lytro.com/) we will certainly not have to wait for a very long time! Unlike anything else you have seen before this camera has a 1 inch Light Field sensor. Which means it captures ALL the Light from all angles. You don’t have to focus at all because everything is “in focus”! You can add out of focus blur effect in post production if you want. That’s it. It is also simultaneously a 2D and 3D camera. You can change the perspective (to a certain degree) in post production too!
Just play with the image sample below. As you can see you can do quite a lot in post production:
And I am pretty sure Sony is working on a camera like that too…would you like an A7 or A6000 camera with such a sensor? Me yes!
UPDATE: There is a great interview at Theverge.com. This is an interesting line:
This is the future. Not the Illum, necessarily, though it’s one of the more exciting cameras I’ve seen in a while. Maybe not even Lytro, though it’s built a huge lead in its nascent industry. But light-field photography — the notion that the future is about turning the complex physical parts of a camera into software and algorithms, that capturing beautiful photos is little more than a data-crunching problem — seems almost obvious. Why capture one photo, from one angle, with one perspective, when we could capture everything? When I can explore a photo, zooming and panning and focusing and shifting, why would I ever want to just look at it?
The dead of classic photography?
0 Responses
For $1,599 it looks interesting, still a prototype like thing in my eyes though. It’s miles better than the previous one and I’m sure that the next will be miles better than this.
More info here:
http://time.com/68862/lytro-illum/
Yep, eventually it should be usable but will always be expensive to get a mediocre final image compared to a traditional camera using the same sensor.
you forgot to add that the digital will never outcompete film.
This is interesting, as was the original Lytro. However, I think it is still a ways off from impacting Nikon/Canon/Sony. The primary reason is that the output from these cameras is very low resolution. VERY low. The original Lytro was a 480×480 image that was resampled to 1024×1024. This one is larger, but still no comparison to ‘modern’ cameras.
That said, the ability to refocus after a hasty shot is incredible, and anyone who says ‘everyone should know how to focus properly’ is kidding themselves, this is a killer feature even for pros.
We always ask ‘What’s next?’ and I do think this shows one possible way forward in photography, its just not here yet.
Change perspective? Since perspective is based on the point from which the photo is taken, I don’t see how this is possible.
It’s extremely narrow based on almost the width of the aperture of the lens itself, so not much to play with at least our two eyes are farther apart from one another so unless you use a huge lens it pointless. Oh and to get enough light and info you would have to take a very long exposure of a still subject. Make no mistake the sample picture was stationary.
While the camera is interesting the trade off in resolution makes it undesirable for me.
Here we go again!
The fact is that the light gathering is limited by the lens and there are a limit to how many pixels you can get on the sensor. Given the way a light field camera works this limits the DR and resolution dramatically; beyond the point of sense!
It is far easier taking pictures the way we do with the equipment we have and it is likely to remain so for a very, very long time, unless they can redefine quantum physics!
Well said !
We can all move on to the next rumor, this horse is officially dead 🙂
Maybe from a professional point of view, you are right, but the mass market has different desires and wishes. The definition of the best image quality is probably quite different for this market. If future lytro like camera’s get 5MP for a “normal” price, than this will sell. Especially with some marketing slogans using 40MP sensor etc.
Looks obscene. Please, no!
Looks interesting…but would have to be in a “useful” format (DSLR) and not toy camera format to be of much use.
In my opinion, these cameras are really not amazing in terms of capturing images. Even in their trailer for the new Illum, they show a skydiver using a gopro (which has a defined purpose). In order to get any sense of refocusing, all Lytros sample images have some sort of object in the way of the rest of the image to give the illusion that refocusing is useful – tree leaves, flowers, a hand, etc. It would have been way cooler if they just focused on building large sensors. Imagine their technology adapted to a 50MP FF sensor – you could refocus images with any lens you wanted. That would be amazing. Of course, the images wouldn’t quite look FF, but it’s an idea. For $1600, there are so many cooler cameras available with higher res, high ISO, video, bokeh, etc
I think these cameras could possibly be the future. I also think that future is still far down the road. One day lenses will be of little value- everything from bokeh and noise will be worked out by software. It’s like television sets. It’s great we have HD and 4K but the real revolution will be holographic technology. One day you’ll flick a switch and be in the middle a jungle right in your living room.
We’ve only begun to scratch the surface with what the future of photography may be. It has many branches yet to be explored.
When I focus on a point, that point doesn’t seem to be all that sharp, meaning its either pretty low resolution, or interpolating between keys (and possibly both).
So still feels likes a ‘single feature camera’ that sacrifices everything for that one feature. Looks good for the web, but not so sure how useful it would be for final prints (resolution being the biggest concern, with sharpness not too far behind).
Yes, might be better of using focus bracketing with a proper high res sensor.
For still objects, yes. But for sports? I think sports photographers will prefer this to their DSLRs. Not now, but eventually.
Looks interesting but with little practical use. The technology has too many limitations and too little advantages in my opinion. Also the fact that they are not releasing much info about the technology makes me think there is something dodgy there. With the previous version for example you could only manipulate the images that contain all captured info on their site and export jpgs through a web-based app.
I have come to think depth of field ( as determined by aperture and focal length) is a rather poor and limited tool for subject isolation and selective sharpness. So I look forward to something of this sort as a new tool offering new creative options.
Looks like no one is fooled into thinking these camera’s will replace our current camera’s. Still they are very interesting and like you all have agreed, they have a purpose, it’s just very narrow purpose :-). Yep you always sacrifice something to gain something else. Marketing tries hard to play that down. Hmm now my super High res sensor is reduced to the resolution of a toy camera but I can capture all the bad focus points along with a few good ones with just one click. I guess if you can’t focus when taking the picture then this is your camera 😉
40Mp means 10M effective pixels or 160M total pixels?
whatever it is, don’t expect too much about its high ISO. the sample shows that the high iso performance is just not more likely than an iphone,
It’s 40 Megarays, not 40 Megapixels. Not sure what that translates to in MP, but I don’t think it’s much.
About 4 MP.
In an interview with the CEO, they referred to the sensor as being a 1″ sensor with an underlying 40MP resolution. So far, reports are saying a final output of an image (or flat file as some call it) is roughly around 5MP in resolution.
Unfortunately, I think the sharpness is too low and the details are missing too many, just like I’m watching a video tape and others are watching 4K blue-ray dvd.
They need to research some sharpening technology to generate simulating details.
It is too much for normal people buy a video tape recorder with $1500.
How sad, I wait for it for 4 years.
I just lost all interest after reading an article in Time about the Illum. Lytro says the camera is made for “creative poineers” and professionals and amateurs looking to stay on the cutting edge of “storytelling photography.” I hate marketing garbage double speak. You mean hypsters with disposable cash who can’t operate outside of auto.
“You mean hypsters with disposable cash who can’t operate outside of auto.”
aka CaNikon users 😉
Yep, lol those Canon & Nikon shooters who probably make up 95% of the pro market don’t know how to shoot. I would have to suggest that Sony are the kings of the “gadget cam”. At ever level they have oddities from the Sony qx10 to everything but the kitchen sink “gadget cam” A99 { well everything bar decent C-AF and good high ISO}
Oh, come on! This is one more useful tool. Photoshop is another and I don’t see it as a “hipster-only” product. And Photoshop is doing what this camera, essentially, is doing – it gives you ability to postprocess a captured moment. Since you are a pro, does it mean that each and every one of your pictures is taken perfectly, because you are totally comfortable with operating a camera “outside of auto”?
Actually I am pretty sure that apps like Photoshop will be deprecated after adoption of light field cameras. I don’t use PS now anyway, although I can see it being useful in some cases.
This idea was brilliant, but never practical….PLEASSSSEEEE….tell me Sony is not wasting valuable time and money on this.
Let Lytro spin their wheels, we need Sony working on:
Autofocus
High ISO
Stabilization
Color reproduction
and
Lenses
Yep, if they are then it’s because the Marketing bone heads are pushing it, sort of like those silly high prices fans with no blades, ohhh ooohh its magic, I’ll pay anything to impress my friends. lol
With this tech you DON’T need autofocus. I hope Sony adopts this tech ALONG with other tech.
Yes, but I believe the point from those who are not impressed is that the resolution is so low that it is just as bad as being out of focus or worse! Since it keeps everything is in low res focus, unlike today’s camera where at least something is composed by the shooter in hi res focus and possibly most everything they want.
I don’t disagree with that. But, given the pace at which storage speed is moving, and the pace at which resolution is moving, the point at which this becomes a viable replacement for AF is not that far off. We’re doing 4k video @30 frames/second right now to UHS-1 SD cards, that’s not even UHS-2.
This tech produces 1 megapixel images. I’ll take the autofocus instead.
This tech will not replace autofocus for anyone except shooters with EXTREMELY low standards. Not now, not in the near future, probably not in the distant future.
Cameras that can take 6 shots, at different focus points in 1/100th of second already exist and do THIS cameras intended job. Those kinds of features will improve and always and forever stay ahead of Lytro.
I’m sorry…it sounds great, and the guy who invented it deserves kudos for thinking creatively, but as a practical application it’s JUST MATH…it doesn’t work.
And Sony only has so much time and money to spend on R&D…if they waste it on stuff like this, we get short-changed.
who needs steam engines when we have horses?
steam engines eventually lead to wars, global warming and self destruction. Some prefer to keep horses 🙂
Yes, but some is the minority and the majority wants professional looking shots for facebook and instagram, mind the word looking.
It is not a coincidence that Google recently added a blur app to Android, in order to make the photo’s look more professional…
Once this camera / technique comes in a smaller package and for a more affordable price, it will catch on for sure.
I’d prefer this to AF. I don’t know about this camera per se, but all things being equal, it’s more sensible to prefer light field focusing than AF.
“This is the future… — that capturing beautiful photos is little more than a data-crunching problem — seems almost obvious.”
-It seems obvious to me that “beautiful photos” are not reducible to “data-crunching”.
“Why capture one photo, from one angle, with one perspective, when we could capture everything?”
-Because “everything” may not be the story I want to tell.
“When I can explore a photo, zooming and panning and focusing and shifting, why would I ever want to just look at it?”
-Because after the novelty wears off, the reality becomes effort, and work.
Effortless viewing will always be more marketable, and enjoyable to the viewer.
______
Seriously, how many photo novelties must pass us by before we realize that novelty always fades away into obscurity. It’s good to have these tools. But I don’t see the masses kicking down the doors of Lens Baby or Cross Processing any longer. Even HDR had it’s moment in the sun… now it looks cheeky when applied to far.
You’re missing the full implications of this. I was shooting last weekend, and even with a decent hit rate, I missed a hundred action shots. With this tech, that NO LONGER HAPPENS.
I get your point. You may have it and I wish you well with it my friend.
My point however, is that If technique is a problem, then I’d prefer to work on my technique, rather than spend more time in front of a computer fixing it.
Seriously, I was a published sports photographer back in the day before anything AF. I still use my AF cameras primarily in manual focus. So I can see your point for you. But I’ll still encourage togs to refine their shooting technique in the field, instead of making up for lacking technique at a desk.
Although I agree with your point regarding technique, the reality is that most people shoot AF. 90% of the people carrying those “big” DSLR’s around them, because they “look professional”, only know one mode and that is auto. I see it with my friends, colleagues and family, the have no clue what ISO or manual focus means, the only technique they know is done by the camera, for them. In that respect, I think that Lytro will come to the mass market in one way or another.
But there is nothing to ‘fix’. The point is that after the photograph has been taken, the viewer can look at it how he wants: stereoscope or flat; shallow focus or deep focus. This will be very attractive for digital magazines IMHO.
I can’t believe everyone thinks this type of tech is not coming to a camera near you… Sure we won’t be shooting with this kind of technology in the near future….But we will…inevitability….
I never print photos and don’t need any more resolution than FullHD (what’s that, 2Mp?), so this seems extremely interesting. Especially given Sony’s unambitious AF, if they ever implement this technology.
“When I can explore a photo, zooming and panning and focusing and shifting, why would I ever want to just look at it?”
That does indeed seem like an interesting question.
Why would one prefer just looking at one picture rather than traveling around in it? Well I can only speak for myself, but when I look at somebody else’s image I am mostly interested in what the author of the image is showing me by eliminating all the things from the recorded scene he didn’t deem important. It’s a proposition, a point of view and it allows me a glimpse into somebody else’s mind, into his thoughts and experiences. As such, while I am always interested in technologies providing me with as much information as possible to be able to rework an image at some later point, I am not sure that light field photographic imagery is all that interesting as a presentation medium. It’s in that sense a little like the debate about 4K or not. To tempt a comparison: for me the relation between a photographic media that contains all information of a particular place at a particular time and a photographic image in the classic sense that only focuses on a very small portion of it is a bit like the relation between let’s say “The Great Gatsby” to “Microsoft Word”.
Bob said: “a bit like the relation between let’s say “The Great Gatsby” to “Microsoft Word”.”
Well said! I like that.
But some of us would rather use Word and write a story instead of simply reading an old out of date story like Gatsby… In essence this technology would give photos a longer shelf life.
Interactive photos seem to be the new thing. Harry Potter, Vine etc. Personally I don’t like DOF as an artistic/visual method at all. In composition the most powerful thing in cropping. Properly cropped image will draw max attention to the main subject. Cropping and the rule of thirds are the way to go. DOF also draws attention in a more subtle way but gives a strain to your eyes as they need to readjust each time you look at a different part of the image. You quickly become frustrated and even disgusted since there’s no detail in blurred parts. It’s just like radically dropping the resolution for the most parts of a high res image under shady pretext of highlighting its subject and still calling it a high res image. If you don’t want me to look at the blurred parts – just crop them out and leave only the subject.
For 650 euro there’s the HTC One (m8) which can do also changing dof after taking the shot :). Maybe in decades this kind of stuff is going to be interesting enough (quality and usefulness)
Most pictures today are being published on Facebook or other online media. Printed high MP pictures are the minority. Many of these pictures could benefit from Lytton technique imho
Resolution doesn’t matter, sharpness is overrated anyways
This sounds revolutionary. Wait… take many photos at different focus distance. when a mouse clicked at a point in the picture, pick the photo which has the sharpest focus at this point. Bingo, you have Lytro Illum.
As long as nothing moves in all the pictures you take.
Irritating gimick. Once I’d played with it for a few minutes I was done.
No thanks
BTW This lytro camera retails for $1500… Good luck with that.
The Lytro camera is very interesting technology, several applications come to mind like security cameras and Google maps street view recording. As a still photographer, the point of focus is one of the first things considered when composing a picture. So this technology while interesting is superfluous for my purposes, and in effect is wasting bandwidth by recording aspects of the scene that are not required. This could be of use to the casual (snap shot) photographer, giving the option of composing and focusing after the fact.
This camera is experimental, a different view with a mouse or double tap in your phone. I’d like to see a parallel technology that would simultaneously take more of the light spectrum than that we cannot see: from infrared passing by visible, and ending in ultraviolet images, in a similar transition. Maybe in the future…
There is no free lunch.
What’s the point of in-focus images when they are super low-res/noisy? We’re not talking about ‘average-smartphone-bad’ IQ. We are talking about god damn awful IQ that doesn’t even look good downscaled for Facebook. That’s also why you don’t even see many hipsters posting their Lytro shots in SNS.
In the future it might well be the standard. Today it’s just a gimmick.
nonsonyalpharumors.com
Click to change focus just reveals out of focus images.
Stupid junk. Boo!
No, its in focus, it’s just so low resolution that its like being out of focus, depending on the resolution you expect, and most real photographers expect high resolution with today’s camera’s. So there you have it, it is what it is. This message has been approved by me.
No, some of it is is out of focus for sure, it’s a matter of depth. The rest is low res, yes.
it sure looks like that.
It looks like a wang.
… you see what you like
I’m sorry guys… but be serious. pictures from these cameras look awful so far… it is not only about resolution… but there are artefacts and other stuff… looks like worst kind of NR…
For me, an important part of photography is about the fun dealing with limitations. If the end goal is that I wear a 360° always-on camera in my holidays, and compose all pictures back at home, this is only one: boring. So no, thanks.
Something is missing!….I know…What does it have inside..Pretty Pics do nothing for me…what is the MP? ISO? you know…the meat and potatoes…
“Flat” files are around 5MP. The 1″ sensor has around 40MP, or 40 million photosites, worth of resolution.
I think this company should instead be focusing on putting this tech on webcams, cellphones and other smaller devices, instead of trying to convince the “pros” that this is the next new thing. And at 1500$, I hardly believe they’ll be able to sell enough quantities. Still, Kudos to them for at least getting the ball rolling. This tech looks fascinating, and I’d be specially interested to see what it’ll be doing in video. The next decade is going to be interesting!
The proof is in the pudding. I am curious to see what people will do with this; we may be impressed sooner than we may think.
Newly developed tech pretty much *always* starts by being introduced in high-end everything. ABS, disc brakes and hybrid in cars, touchscreens on smartphones, autofocus and digital sensors in cameras, the list is endless. I can;t think of a single example of blistering new tech being introduced at the low end.
I suppose this allows for quicker recovery of development costs. Usually new tech has a limited capacity for mass-production, so it’s easier to recoup USD100 (to name a random number) per device if that device is a top-end USD1500 camera than if it is a USD25 webcam, etc.
I like the idea, but I don’t like the follow through. The technology is neat, but it is very proprietary and there is no standard to follow. This in itself makes it a very limited application.
What I mean by this is that in order to take advantage of its capabilities you are forced to use only their software. If others release similar cameras, we will be limited to using just their software to process our images no matter how bad or good their editor is. This has had a negative affect on Sigma’s Foveon and has in the past on other sensors that used unique bayer type filters. The new Fuji’s are the exception but have taken a few trial and errors before the interpretation of the images started to meet expectations. These light field cameras take this several steps further, so you can imagine the complications.
Now if these light field cameras become popular enough and the camera industry can agree on a universal standard for image data interpretation which would open up the door for using different kinds of editors and plug-ins, then we may have something here. But I don’t see this happening any time soon since each company selling one of these wants to have some capability their camera can do that the others can’t.
Great ! that means no more focus stacking for macro ! no more focus.. capture .. for 7-10 times !
take one shot, then all you have to do is to move the mouse a little then save a photo and keep doing this for several times then stack the photos in photoshop
damn .. I love this idea
You don’t even need to do that. From the looks of it, you can even choose to have an image where the everything is already in focus. So no additional jobs of focus stacking is needed.
I just saw that .. one click and baam .. all in focus
this is crazy
That’s not possible, because you are still limited by diffraction.
Use the 7s at a very high iso and stop the lens down to highest f stop will yield better quality and more dof than this.
Alternatively use a 1/2.5 ” sensor point and shoot like the HX400 for more depth of field , higher quality, and extreme versatility as well as a much lower price tag.
Good point Barry! Better hide for a few days… the Marketing folks from this company are going to try to find you and quite you, know what I mean aye know what I mean 😉
Exactly, an A7S will do better with 3 stops higher ISO. A dedicated chip could temporary store the 8 images and then stack them. The lens would be similar but instead working at the same time the 8 seperate parts would be delayed, because they would have no seperate sensor, all of them would use the main FF sensor -> no simultaneous shooting, but one-by-one, turn the ISO up and you won’t loose time.
I don’t know if you’ve heard but there’s this thing called diffraction that blurs photos more and more when the f-numbers are increased.
Also, I’m not sure if you know that stopping a lens down puts everything in to focus. The light field method still allows for selective focus.
That’s without even talking about the “3D” feature, which I don’t personally care about much, but is still a neat option to have.
You seem to need to invent total nonsense reasons for the sole purpose of dissagrreing.
Diffraction on a 12 mp image?
On a 12mp image, diffraction starts to set in around f/11 and by f/16 it is fully set in for all wavelengths.
Unless you’re in the wide angles, even f/16 is not going to help.
And what is your response to at least the selective focus the lytro can do?
They should’ve called it the Esper.
“Enhance 224 to 176. Enhance, stop. Move in, stop. Pull out, track right, stop. Center in, pull back. Stop. Track 45 right. Stop. Center and stop. Enhance 34 to 36. Pan right and pull back. Stop. Enhance 34 to 46. Pull back. Wait a minute, go right, stop. Enhance 57 to 19. Track 45 left. Stop. Enhance 15 to 23. Give me a hard copy right there.”
[On a completely unrelated note, I stopped getting e-mail notifications of new SAR posts a few days ago. Anyone else had that happen to them?]
Am i correct that this camera is just always stepped down and you use an Software to set the focus?
Isnt it just the same then Shooting at f/16 all the time and use a cheap plugin to set focus and blur the rest?
nope. it’s a complete different technology. you can’t even imagine what the camere is doing, because it’s not what your eyes are doing. and I don’t think this would work with conventional lenses btw
So to sum up.
This Camera have an 40 Megapixel Sensor and a Special Lense in front of it which splits up the Light in a way you can take 8 Photos at one time with 5 Megapixel each.
So you have all disadvantages of an 40 Megapixel 1″ Sensor, but only 5 Megapixel and the only advantage is that you can refocus afterwards because those 8 Images are stacked? Is that all the magic?
Not to think about the influence of the image quality of this lense. Its only one single lense, no chromatic correction, no distortion correction and so on.
This totally sums up our Generation.
– To stupid to take photos
– No demand on Image Quality
No wonder Google is trying to get this into Smartphones, this is where this technology belongs.
So it’s just software based focus stacking like on the Mokia Lumias? That sucks. An A7S with a good image processor and buffer could take 8 images in a few ms and stack them… You don’t loose shutter time on the lytro, but utilising 12.5% of the sensor is a joke. The sensor from the A7S and a next gen chip could do this much better (doing the trick with shutter time, not with the surface area garbage)…
Smartphones… On Sunday I was at the beach in Luanda with my A7r & 55 f1.8. A friend had his i-Phone 5S. Some guys arrived with their beach boards and started the fun. Whilst the A7r was trying to focus my friend took the shots and these were much nicer. Even though the i-phone 5S IQ may not beat the Sony A7r, I recognise that in some circumstances a smartphone is the best option… It is also much smaller than the A7r. Smartphones can too produce nice pictures… we should not be sectarian. Afterall, it is the more result that counts, less the way we got to it. Any technology is good to have !
key point is: smartphones are only useful “in some circumstances” i.e. under a limited set of conditions and IQ requirements. Personally, I don’t shoot photos with a smartphone and never going to. At least until they build an Android smartphone around RX100.
Rx100 is pretty much like a phone camera. Slow start up… electronic zoom, slow lens, tiny sensor… etc.
Your problem was you didn’t learn how to use the a7r in all conditions. That’s not a strength of the iphone.
If you’re such a hard core purist you must only shoot film and not crop or adjust in post.
Sony should make a Play Memories app that does almost the same thing as this.
Basically, take several photos in rapid succession, but change the f-stop each time. It wouldn’t work with motion, but it would make capturing images for stacking faster – similar to HDR bracketing, but for f-stop.
changing F-Stop isn’t a good idea, cause that affects the shutter or ISO value. AF is fast enough nowdays. There could instead be a mode, that takes a few pictures, in rapid succession, each with focus point shifted further and further out to infinity. Of course the camera would have to calculate the depth of field at a given f-stop and focus distance, on the fly.
So after reading some stuff on their website, the lowest shutter speed is 1/250. That tells me that this camera, more than likely, takes a few photos in a rapid succession. Say 5 photos, which is probably more than enough for a 1″ sensor, at 1/250th. That would come out to 1/50th total shutter speed over a span of 5 photos. Seems alright.
But that also tells me that with it’s f2.0 lens at 1/250th, this camera can take pictures in daylight as well as well lit interiors.
If Sony were to introduce a simple option in their cameras, where it takes a bunch of pictures and shifts the focus point, it would essentially do the same thing as this camera, with a much better quality.
This camera does not need to take multiple pictures to get all the light data, it is simply spread the information out over the entire sensor, so it’s more like taking individual pictures on different sections of the sensor (its more complex than that but that’s the basic idea). Low shutter is more likely due to the Sensor pixels being so small and needing to gather more light per pixel to get decent results.
Don’t know about “more light per pixel” since exposure is exposure, and ISO goes up to 3200.
Using your logic, the only way that it could capture multiple images at different focus points using a single exposure, is if the sensor is broken down into multiple parts(4, 9, 16, etc), and each part being at a different distance to the rear element, than the other. 9 Sounds about right, since that would yield about 4.5mp per sub-sensor. However, it has only 1 lens, which means that it will be a crop of a specific area under a different focus point, rather than just a different focus point.
A stacked sensor, with multiple photosensitive layers, is much more probable, but then you’re losing green light intensity much more than any other color, at every subsequent layer.
The only feasible option is to take a bunch of pictures in rapid succession, and either move the sensor or move the lens focus point to a predetermined distance, while In camera processing will resample to 5mp and process the image.
I haven’t seen any reference to splitting the sensor into multiple areas anywhere or stacking layers, so i’m still standing by the focus stacking technique.
Stan,
Look up “Plenoptic” online. Adobe is one of the original pioneers of this technology. The sensor actually has hundreds of micro-lens, each covers about 100px. Each of these micro lenses focuses on a different part of the light field, computation afterward allows these images to be recombined to create the lightfield image that Lytro show in their web viewer.
The main lens in the Lytro camera is just used to capture the light field and put it in the right place in the camera, which is why it can be manufactured without the expensive aspherical glass that most modern lenses use. A normal sensor just captures one plane through the a lightfield, which is the focal point. Tilt-shift lenses literally reposition the lightfield which is why you are able to achieve a focal plane that is non-perpendicular to the image. It’s a really fascinating technology, but it does not capture multiple images in sequence. It captures the entire lightfield at once.
What I think you’re missing is that inside a camera, there is not just a single plane of light, that’s how modern cameras represent the image, but in reality the light field exists a 3 dimensional collection of light, the camera is just capturing one plane of it in focus. The Lytro camera is like having a bunch of miniature cameras each aimed at the 3D lightfield capturing a different part of it, all at once.
I think i get it. As soon as you said microlenses used to focus, a lightbulb went off.
And thanks for the tech name. That helps!
Adobe might be the commercial pioneers of this technique. .. at best they may have patented something.
AFAIK It doesn’t take pics in rapid succession using the image sensor, instead it captures various light rays’ positioning info in the scene under different angles and then uses the sensor and Qualcomm 800 to render a particular image using a subset of that info and proprietary software. The image itself does not exist on its own, It’s kind of rendering a scene in Maya – the result is different each time depending on camera pos and various given parameters. Lytro is about software processing of whatever rich scene info you can get using various types of sensors. This tech is based on 20 years of R&D research of the company’s founder. Key point: the 11 MegaRay version could only produce 5 MP images – not sure about the Illum with 40 MegaRays. He also said that the F2.0 8x zoom lens are possible because it contains fever glass elements compared to conventional DSLR lens since it does not need to correct for various image imperfections, distortions and chromatic aberrations
Interesting.
On a side note, MegaRay sounds like something out of Pokemon or one of those Japanese cartoons. LOL
Everyone is debating back and forth, meanwhile I am enjoying how cool this technology is in the demo.
We already enjoyed the demo, so now it is time for discussion on professional applications. So far it seems to have a long way to go for that.
this is an interesting concept tho that would mean shifting the skills from photography to post possessing which isn’t really photography now is it? lol ….This is basically making aperture control irrelevant when taking a picture….
minipixels.net | Automotive Photography + Design
The point is that there is no post-processing. The viewer decides how to view the image. I’m sure that some people who spend more time with the computer than taking photos will hate it. That’s my take.
“shifting the skills from photography to post possessing which isn’t really photography now is it”
Tell the 2 billion (or whatever :p) users of Instagram this. Real photography is arguably what most people do, not what a tiny elite of [medium format shooters/developing their own film/owning only MF rangefinder cameras/all of the above] decides what is “real” and what is not.
(and of course developing film and printing it is just as much post-processing as using software is…. so PP has been part of photography since at least the invention of the negative)
Shouldn’t talk about other people’s skills unless you manual focus action and shoot with film.
I think you are suggesting that photography involves having only one opportunity to capture an image in it’s final form? Post processing (dodge and burn, multi-exposure, etc) has always been an important part of ‘real’ photography. Even if the entire scene were able to be captured in three dimensions + time (so now 3D video) with infinite resolution and zero noise… a good photographer would still pull out an interesting and thought provoking image by picking the right moment in time,the right angle, focal point(s), amount/style of bokeh, introduction of noise, etc. while I, on the other hand, would probably not do so well because I am probably not all that great of a photographer.
In short, who cares if the result of aperture adjustment is achieved at the moment of capture or later? Sure, it’s harder to get it right (most impact of the final image on the viewer) when you only have one chance but it takes artistic knowledge and skill either way.
awesome! while we’re at it, can we also get rid of enormous zoom lenses by using 50+ megapixel sensors combined with Nokia PureView technology, thanks
If only it were that simple!
Nokia’s Pureview is pretty much a resampler algorithm. You can take a 38mp image or an 8mp image with zoom capability.
That allows for a 2x digital zoom, without any image quality loss. You can do the same thing in post processing on any camera. A 24mp image, with 2x digital zoom, will yield a 6mp image. A 36mp image, 2x cropped, would be a 9mp image.
On top of all that you get the benefits of a high ISO capability of any large sensor. Suddenly that Nokia Pureview doesn’t look so awesome.
Sure, you can crop any image further, and then resample it back to a higher resolution, to get smoother gradient. But why would you want to crop a high res image anyways, when you can keep it, at the disadvantage of using a bigger lens.
Ok then maybe we need something like 500+ MP sensors 🙂 Surely we don’t need to keep the original resolution of that, and we would be able to zoom in even more.
To do what you’re saying, even a FF 500MP sensor needs a f/1.6 lens to avoid diffraction. Getting a lens that fast to resolve over 750 lines/mm would be where the fun is for the designers.
The revolution is only allowed to go half way, apparently. Arguably 90% of photographers are satisfied with digital replacing film. But apparently light field cameras are not allowed to replace single plane sensors! “Real photographers use AF”. Sound familiar?
When this tech reaches FF sensors and would allow resolution and flexibility of today’s DSLRs or yesterday’s SLRs, then we will talk.
While this is a cool tech, I can see it making to P&S cameras first. I probably will not make it to cell phones, at least its yet to be seen with Nokia’s project.
But this could be something that would revive P&S cameras, since those are not as limited in size as cell phones are.
This could as well be something like a BSI sensor, which benefits high density sensors way more that it does low density sensors.
Arguably, this is because 90% of people are stupid and perpetually mistaking current market offerings with the laws of physics.
Standard camera designs are not in trouble any time soon. All the Lytro technology does is allow for selective defocusing, nothing more and nothing less. It’s equivalent to shooting a series of high-f/stop images in a grid pattern, then recombining them such that objects at a desired distance are aligned in the source images. Everything else will not align perfectly, giving out-of-focus blur.
There are some neat things that can be done with these data, but there are also major disadvantages. Resolution is decreased as the possible focal ranges is increased, meaning that to double the depth of field of a traditional camera you’d need to halve the resolution. With the existing Lytro design the main lens cannot be focused, meaning that the camera can never focus closer than designed (probably the hyperfocal distance). Diffraction is still a problem, so this cannot be used to overcome resolution limitations of macro subjects.
Even if the lens doesn’t focus itself, you can attach macro ring in front of it.
>cannot be used to overcome resolution limitations of macro subjects.
Yeah. But at least, you will now be able to shoot live moving insects without freezing or killing them.
You’d probably need a specialized macro ring to ensure that the exit pupil size of the lens does not change. What’s captured by the camera in raw form is an array of images of the exit pupil, and an increase in its size would cause the sub-images to overlap and contaminate each other. A decrease in exit pupil size would probably be ok, but you’d waste sensor area and not be able to “re-focus” as much.
Guess what. Front mounted converters don’t change lens exit pupil.
Ah, you’re right! Thank you.
“Dead [sic] of clsssic photography?”
Badly placed shallow dof, is not what makes all bad photos.
This would be a nice feature for those once in a life time moments. Where for some reason the camera focused on the background instead of your subject, but I wouldn’t want to rely on it. I like the feeling of creating a nice picture. Its like a crutch for someone who has no clue about cameras and how to use them. Other then just push a button and know your pictures will be ok with a push of another button. Why not get rid of all photographers and just have robots pushing 1 button on the lytro camera. Robotography. Lets not come to the point in life like the people in the animation wall-E… “operate manwell” Clueless/Dumb.
In case of shooting this technology is P&S, It does not allow you to use different lenses for their special picture, does not allow you to use special techniques like multiexposures, etc, etc… Just Point to what you like to be fixed and shoot.
BUT in case of USING taken pictures this technology leads to heavy postprocessing of EACH shot. Which is not good for P&S shooters.
This makes these cameras to be just a technology toys… Serious photographs need more control over shooting process, while P&S shooters seed simple importing without postprocessing for sharing, printing, e.t.c.
At some point, we will see plenoptic cameras with traditional IL mounts. Why? Imagine you’re shooting landscape with Sigma 35/1.4 lens in lowlight. You have to stop it down for about f/11 to get enough DOF and raise ISO, and you are simply carrying extra bulk of f/1.4 lens. With plenoptic camera, you can shoot wide open. Even more, plenoptic cameras give more options to correct aberrations in postprocess. An option to get use of these many megapixels.
Umm high isos landscapes because you were too lazy to have a tripod around… seriously? And you think a lytro won’t give noise or experience shake?
All your problems can be solved by using a tilt shift lens and a tripod. People have been doing this since at least 1901.
Umm autofocus because you were too lazy to manually compose… seriously? And you think an AF camera will be fast enough to be quicker than MF and won’t be a drain on the battery?
All your problems can be solved by using cheap Manual Focus film only cameras. People have been doing this since at least 1800s.
I don’t think you can read. I wasn’t making a case for either OP’s claim or lytro’s case.
Also what crack are you talking with “manual focus film only” cameras from 1800? The Scheimpflug principle was first documented in 1901/1904.
TS lens won’t solve my problems, cause i demand tilt in-body.
and while shift does help,
TS helps, but doesn’t allow still to fit a insect, even a cat into area of focus, unless you flatten them in before.
tripod doesn’t stop clouds and trees from moving.
Tilt/movements in body? It doesn’t make any difference to the end effect. You can virtually achieve this by fitting the tripod on to the lens instead of the body.
Also a t/s lens has the potential to freeze motion because you won’t need as big f numbers. You can still use high ISOs as well you know?
And what do you mean by cats and insects? You were talking about landscape photos.
5MP is still too low. I’ll wait for V4 or V5, before I put any hard cash down.
can we shoot one light or multiple light with this camera?