If you can read this, either the style sheet didn't load or you have an older browser that doesn't support style sheets. Try clearing your browser cache and refreshing the page.

(Phys Org2)   Engineers build 50 GIGApixel camera. Think about it... Both your mom's buttcheeks in one picture   (phys.org) divider line 34
    More: Spiffy, spectroscopy, field of view, PhysOrg, Space Telescope, synchronizations, objective lens, Michael Gehm, pileup  
•       •       •

2899 clicks; posted to Geek » on 20 Jun 2012 at 9:16 PM (2 years ago)   |  Favorite    |   share:  Share on Twitter share via Email Share on Facebook   more»



34 Comments   (+0 »)
   
View Voting Results: Smartest and Funniest

Archived thread
 
2012-06-20 09:21:00 PM  
My first thought: That would make an awesome spy camera for U2's and drones.

FTFA: "The team's research was supported by the Defense Advanced Research Projects Agency (DARPA)."

Still awesome.
 
2012-06-20 09:28:23 PM  
That might be the most juvenile headline I ever LOLed at.
 
2012-06-20 09:29:36 PM  
I don't know why, but I love a good your mom joke. And this is a good one.
 
2012-06-20 09:33:28 PM  
Creepy guy that invented the camera so he can maintain the restraining order distance while still being able to take pictures of kids at the park.

i48.tinypic.com

/I kid. I'm sure he's a nice guy
 
2012-06-20 09:38:54 PM  

SilentStrider: I don't know why, but I love a good your mom joke. And this is a good one.


My reactions to headlines like this assure me that I'll never grow up.
 
2012-06-20 09:40:44 PM  

Flumenos: My first thought: That would make an awesome spy camera for U2's and drones.

FTFA: "The team's research was supported by the Defense Advanced Research Projects Agency (DARPA)."

Still awesome.


ELEVATION!

WOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
 
2012-06-20 09:52:41 PM  
When will we have this commercially and we can all have the Camera of Omens with sight beyond sight?
 
2012-06-20 09:57:35 PM  
If you could display a 50 foot by 50 foot, 50 gigapixel image, the distance at which individual pixels would be distinguishable with my 20/20-corrected vision would be close enough for my erect penis to be pressed against or penetrating the display.
 
2012-06-20 09:57:38 PM  
Ok +1 subby. Nice.
 
2012-06-20 09:59:29 PM  

Whatthefark: Creepy guy that invented the camera so he can maintain the restraining order distance while still being able to take pictures of kids at the park.

[i48.tinypic.com image 400x598]

/I kid. I'm sure he's a nice guy


It wouldn't surprise me to find out that guy has a secret lab built inside a volcano.
 
2012-06-20 10:00:34 PM  
I LOL'ed.
+1
 
2012-06-20 10:09:40 PM  
synchronizing 98 tiny cameras in a single device

Megapixels don't matter as much as lens quality, tiny cameras == tiny lenses == tradeoffs.

The camera's resolution is five times better than 20/20 human vision over a 120 degree horizontal field.

A FOV of 120 degress = a 12mm focal length (on a full frame camera with a rectilinear lens).

So it's really wide, but that's what I would expect from something made from a bunch of cameras/lenses slapped together. They're essentially making a huge panorama across the X and Y axis. And probably only grabbing the sweet spots at the center of each camera/lens (vignetting and chromatic aberrations occur more frequently at the edges of lenses).

So although this is interesting, it's not much of an advancement. Tech wise, it's a no brainer. The concept is no different than a regular panaroma (shoot a bunch of images, stitch them together). They're just parallelizing the photo taking.

// The biggest panorama I've made was of 80 images. 80 * 8mp => 640 megapixels ... a little over half a gigapixel. And that's with one camera panning over the Y axis only. Another vertical row of 80 photos would have gotten me to 1.28 gigapixels.

// article is boring with specs on the cameras. How many megapixels per camera? Focal lengths of the lenses? Fstop range? etc?
 
2012-06-20 10:10:32 PM  
We're gonna need a bigger memory card!


also awesome headline.
 
2012-06-20 10:11:17 PM  
s/with/without/
 
2012-06-20 10:13:26 PM  
PS, welcome to the danger zone

sphotos.xx.fbcdn.net

// Sorry, there's no highway to the danger zone. Only dirt roads.
 
2012-06-20 10:20:34 PM  
Found a better article.

http://www.livescience.com/21066-supercameras-capture-detail.html

Basically, they have one huge lens (essentially a collector), and the 98 sub cameras get their light from the collector.

Also, it's only capturing slightly over 1 gigapixels. The 50 megapixel statement in the article is future looking.

The new camera has the potential to capture up to 50 gigapixels

// completely not impressed anymore
 
2012-06-20 10:34:59 PM  
i479.photobucket.com
 
2012-06-20 11:26:39 PM  
FTA: By synchronizing 98 tiny cameras in a single device...

Sounds familiar.

img38.imageshack.us
 
2012-06-21 12:21:06 AM  
The researchers believe that within five years, as the electronic components of the cameras become miniaturized and more efficient, the next generation of gigapixel cameras should be available to the general public.

I kinda doubt that. There's a quantum issue in packing pixels into imaging chips, and we're bumping up against that now. The pixels still have to be larger than the wavelength of the light they receive. So larger CCD's are about the only path to larger commercial products.
 
2012-06-21 01:36:54 AM  
Pictures, or it didn't happen.
 
2012-06-21 02:06:16 AM  
For once I truly Laughed OL.

/with apologies to FARK favorite, "Meow Said The Dog"
 
2012-06-21 02:14:06 AM  
users.rcn.com
 
2012-06-21 05:08:36 AM  
Vesper approved.
 
2012-06-21 08:39:19 AM  
Oh good, it's going to be used to help kill people without trial or evidence.
 
2012-06-21 09:07:18 AM  

lordargent: Megapixels don't matter as much as lens quality, tiny cameras == tiny lenses == tradeoffs.


Yeah, yeah, and I'm sure the bokeh sucks.

So although this is interesting, it's not much of an advancement. Tech wise, it's a no brainer. The concept is no different than a regular panaroma (shoot a bunch of images, stitch them together). They're just parallelizing the photo taking.

You know what else is boring? Today's supercomputers! I mean, where's the advancement in throwing 10,000 processors together and getting them to work simultaneously on the same problem? You could just as easily, in fact more easily, do it on one processor, just taking 10,000 times as long.

"Just" synchronizing 98 cameras and synthesizing a composite view in near-real-time IS a big deal, big enough to earn a slot in Nature. No offense, but I think I trust Nature's anonymous reviewers more than one anonymous snarker who hasn't even read the paper.
 
2012-06-21 09:09:58 AM  
It's nice that they will be able to capture your moms ass in high resolution subby, but it's focal length and distance to the subject that determines what is in frame. Not mega/gigapixels.
 
2012-06-21 10:34:55 AM  

Flumenos: My first thought: That would make an awesome spy camera for U2's and drones.

FTFA: "The team's research was supported by the Defense Advanced Research Projects Agency (DARPA)."

Still awesome.


What I don't understand is how they handled the optics. Even pro grade camera lenses aren't capable of much over 50 megapixels no matter what sensor you put behind them.

lordargent: synchronizing 98 tiny cameras in a single device

Megapixels don't matter as much as lens quality, tiny cameras == tiny lenses == tradeoffs.

The camera's resolution is five times better than 20/20 human vision over a 120 degree horizontal field.

A FOV of 120 degress = a 12mm focal length (on a full frame camera with a rectilinear lens).

So it's really wide, but that's what I would expect from something made from a bunch of cameras/lenses slapped together. They're essentially making a huge panorama across the X and Y axis. And probably only grabbing the sweet spots at the center of each camera/lens (vignetting and chromatic aberrations occur more frequently at the edges of lenses).

So although this is interesting, it's not much of an advancement. Tech wise, it's a no brainer. The concept is no different than a regular panaroma (shoot a bunch of images, stitch them together). They're just parallelizing the photo taking.

// The biggest panorama I've made was of 80 images. 80 * 8mp => 640 megapixels ... a little over half a gigapixel. And that's with one camera panning over the Y axis only. Another vertical row of 80 photos would have gotten me to 1.28 gigapixels.

// article is boring with specs on the cameras. How many megapixels per camera? Focal lengths of the lenses? Fstop range? etc?


Thank you. Now it makes sense.
 
2012-06-21 10:52:04 AM  

lordargent: synchronizing 98 tiny cameras in a single device

Megapixels don't matter as much as lens quality, tiny cameras == tiny lenses == tradeoffs.

The camera's resolution is five times better than 20/20 human vision over a 120 degree horizontal field.

A FOV of 120 degress = a 12mm focal length (on a full frame camera with a rectilinear lens).

So it's really wide, but that's what I would expect from something made from a bunch of cameras/lenses slapped together. They're essentially making a huge panorama across the X and Y axis. And probably only grabbing the sweet spots at the center of each camera/lens (vignetting and chromatic aberrations occur more frequently at the edges of lenses).

So although this is interesting, it's not much of an advancement. Tech wise, it's a no brainer. The concept is no different than a regular panaroma (shoot a bunch of images, stitch them together). They're just parallelizing the photo taking.

// The biggest panorama I've made was of 80 images. 80 * 8mp => 640 megapixels ... a little over half a gigapixel. And that's with one camera panning over the Y axis only. Another vertical row of 80 photos would have gotten me to 1.28 gigapixels.

// article is boring with specs on the cameras. How many megapixels per camera? Focal lengths of the lenses? Fstop range? etc?


I came here to say this.
 
2012-06-21 10:52:30 AM  

Flumenos: My first thought: That would make an awesome spy camera for U2's and drones.

FTFA: "The team's research was supported by the Defense Advanced Research Projects Agency (DARPA)."

Still awesome.


My first thought was sticking this sucker at the end of a super giant telescope orbiting the Earth and seeing if it gives us any better pictures of whatever's out there.

Of course, I know next to nothing about cameras and telescopes so my thought is probably naive.
 
2012-06-21 11:00:14 AM  
Enhance.
Enhance!
ENHANCE!
 
2012-06-21 11:08:41 AM  
Perfect for the stalker from a long distance away.
 
2012-06-21 01:21:25 PM  
jfarkinB: "Just" synchronizing 98 cameras and synthesizing a composite view in near-real-time IS a big deal, big enough to earn a slot in Nature. No offense, but I think I trust Nature's anonymous reviewers more than one anonymous snarker who hasn't even read the paper.

TL/DR version: It's not rocket science

Tolstoy version:

People have synchronized multiple cameras before, and there's already software to stitch together panoramas. These are not big achievements, more like baby steps. The only difference is the scale and speed, but this isn't something that's very difficult to scale or speed up.

The synchronization and stitching might be technically difficult, but they are not conceptually difficult.Conceptual difficulty impresses me far more than technical difficulty.

Technical difficulty makes me say "how did they get around this"

Conceptual difficulty makes me say "how did they think of this"

This project makes me say.

How did they get the trigger signal to 98 cameras?
How did they work out focusing a single lens onto multiple smaller lenses.
How did they work out culling and merging in real time.

It doesn't make me say

How did they think of synchronizing 98 cameras together.

The first time someone synched multiple cameras together was conceptual achievement. But once that was done, it was just a technical question of how do you synch more cameras, or bigger cameras, how much ram do you need, etc. All technical details, but no truly brilliant conceptual leaps.

[synchronized cameras, used for the matrix circa 1999]

files.petapixel.com

They even state in the interview that their most challenging problem was acquiring hardware.

The development of high-performance and low-cost microcamera optics and components has been the main challenge in our efforts to develop gigapixel cameras,

I'm sure there were a lot of technical issues that they had to work out (but to people that work in that field, these aren't huge hurdles). But from a high level conceptual view, it's not that difficult of an idea, the devil is in the details.

// Also, I have to admit that the headline made my expectations high as well. I expected to see a 50 gigapixels on a single CCD or something like that. Instead, I get synchronized cameras that take 1 GP images (and could scale to 50 ... if they could get their hands on the hardware).
 
2012-06-21 01:30:41 PM  
jfarkinB: "Just" synchronizing 98 cameras and synthesizing a composite view in near-real-time IS a big deal

Also, I'm not so sure you read the nature article either.

Their interval speed is "three frames per minutes".

So about 20 seconds per photo, not exactly "near real time".

// if you haven't guessed by now ... camera nerd :D
 
2012-06-21 01:35:14 PM  

MooseMuffin: SilentStrider: I don't know why, but I love a good your mom joke. And this is a good one.

My reactions to headlines like this assure me that I'll never grow up.


Thirded. An unexpected "Your mom" joke broadly directed gets me EVERY time.
 
Displayed 34 of 34 comments

View Voting Results: Smartest and Funniest


This thread is archived, and closed to new comments.

Continue Farking
Submit a Link »






Report