Slashdot Log In
Largest Digital Photograph in the World
Posted by
CmdrTaco
on Tue Nov 16, 2004 09:46 AM
from the thats-a-lot-of-bits dept.
from the thats-a-lot-of-bits dept.
thrill12 writes "Dutch research institute TNO has unveiled what it believes is the largest digital photograph in the world. The image contains 2.5 gigapixels or 7.5 gigabyte worth of data. It is composed of 600 single images shot by a computer-controlled pan-tilt unit in 7 second intervals. Afterwards, all photos where stiched together (compare: panorama tools) using the capacity of 5 high-end pc's in about 24 hours time."
This discussion has been archived.
No new comments can be posted.
Largest Digital Photograph in the World
|
Log In/Create an Account
| Top
| 318 comments
(Spill at 50!) | Index Only
| Search Discussion
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
(1)
|
2
(1)
|
2
New Title: (Score:5, Funny)
Let's rewrite that intro shall we?
Most Boring Picture Ever Taken
Dutch research institute TNO has unveiled what it believes is the most boring picture ever taken. The image contains 2.5 gigapixels or 7.5 gigabyte worth of pictures of the roof of some office park. It is composed of 600 single images shot by a computer-controlled pan-tilt unit that was incapable of actually viewing anything of any interest to anyone. Afterwards, all photos where stiched together using the capacity of 5 high-end pc's in about 24 hours time. Three graduate students died of boredom; services will be held somewhere exciting, like a morgue. Never have so many, downloaded so much, for so little...
Re:The Big Picture (Score:5, Funny)
Scrolling seems to be a bit of an issue though.
KFG
Groan (Score:5, Funny)
(Last Journal: Thursday December 08 2005, @04:33PM)
Re:Groan (Score:5, Funny)
(Last Journal: Thursday June 05 2003, @09:57AM)
And as a followup: (Score:5, Funny)
sigh (Score:5, Funny)
Re:sigh (Score:5, Funny)
KFG
Wow (Score:3, Funny)
Can we take it again? (Score:4, Funny)
(http://seankelly.biz/)
Now we know (Score:3, Funny)
(http://slashdot.org/)
Mirror (Score:3, Informative)
Disappointment. (Score:5, Interesting)
(http://www.danm.net/)
What a waste (Score:3, Interesting)
(http://www.evanagee.com/ | Last Journal: Tuesday November 25 2003, @02:20PM)
Eitherway, I can just see the MASSIVE, high resolution billboards now...
Legit? (Score:3, Interesting)
(http://www.piranhaa.net/)
In case of slashdotting (Score:3, Funny)
Bah... (Score:5, Interesting)
(http://www.bustedskull.com/)
For all non-photo geeks out there (Score:5, Informative)
(http://www.wyattmcguire.com/)
I'd like to take this time to point out the lunacy of the 'megapixel' ratings for cameras
A number determined from the multiplication of length and width in pixels of an image has about as much to do with the quality of a picture as the size of your passenger cabin has to do with the speed of your car. Yes, you can print larger pictures without seeing pixels if you have a higher megapixel count, but chances are it's not the resolution of your photos that you'll notice.
A major factor in the quality of any image is the quality of the optics used to take it. That means the lens, the glass used to focus and point the image onto the sensor. Quality glass, such as low dispersion glass (I'm preferential to Canon's "L" glass) will create images with sharp edges, crisp focus, and good bokeh. Use cheap glass and you'll get the opposite. Effects like soft focus, purple halos, light leaking, and distortion will all still be present if you use poor optics, no matter what the MP rating. I wonder how many people have upgraded from a 3 mp to a 4, 6, or 8 mp camera and still found lackluster results.
My point, a camera has many more features that determine quality than just the megapixel rating, when you choose one, consider these as well and you'll be happier. And here's a plug, dpreview.com does some awesome camera reviews (I'm in no way affiliated with them).
Canon-L fanboys who know nothing of optics (Score:4, Informative)
Bokeh comes from the number of aperture leaves and their shape(there are some non-straight-edged aperture leaves). It has -absolutely nothing to do with the quality of the glass-.
Furthermore, Canon's $60 50mm/1.8 is plasticky, cheap, blah blah- but it's just as sharp as the faster, metal (heavier) L-series lens, and it doesn't suffer from the mild barrel distortion the L-series lens does. It has fewer aperture blades, so bokeh is not as great- which is pretty much the only reason pros buy the L version. Consumers buy it because they want a red ring around their lens and they don't want to be caught dead with a plastic lens.
You can stare at lens charts until the cows come home and argue about image quality. The L-lenses are slightly better in most image quality categories since they do generally use the very best of Canon's technology, but their chief advantage is that they are built with stronger but heavier materials, aimed at professional users who don't mind that the body is thick metal. Phil Greenspun claims he's dropped his 70-210/2.8 IS on the floor and it worked fine. I'm not about to try with mine, but I can tell you that the thing is built like a goddamn tank, and designed to be modular for easy servicing. Even the tripod mount screw is replaceable...
If I want to see Delft... (Score:3, Funny)
First stitching, then tiling (Score:5, Interesting)
(http://www.schaik.com/)
Flash based zoom/pan/tilt viewers do the same thing. A bit more advanced, but you download only the part that is currently in view. Even when you open a PDF in your browser, just the page in view is downloaded. And think about those huge video walls.
So, the funny part is now that you take many, many pictures, then use a lot of processing to stitch the results together, and then cut it into tiles again to display the resulting image. Wouldn't it make more sense to put some more effort in that robotic camera control device and make that so accurate that it can take the pictures, still touching, but with zero overlap? That would be cool!! I suspect that making the high precision optics for such a camera would be really, really expensive. Which is probably why TNO did it the way they did.
Hrm (Score:3, Interesting)
Um, no. (Score:5, Informative)
It's true that the file size of our imagery is smaller than theirs, as we use Mr. Sid format for better compression, but our pixel count leaves them in the dust.
I don't believe this image is in any way extraordinary or special - pretty much every local government across the country maintains digital imagery of their jurisdiction that is comparible in resolution.
11.3 Gpixel in my research lab (Score:5, Interesting)
(http://www.dellamea.it/)
Please be very kind with our test server: http://www.telemed.uniud.it/eslides/ [uniud.it].
(anyway, I never thought this kind of things could become a news item).
not impressed. (Score:5, Interesting)
(http://prototypecreative.com/ | Last Journal: Thursday July 12, @11:52AM)
exposure started june 2002 and ended early november 2003.
i used MacOSaiX to put it together on a two year old powerbook, and it took about 12 hours.
it's not seemless, but the mosaic effect is cool.
How about the inverse? (Score:3, Interesting)
Anyone here have suggestions?
Mine's bigger (Score:3, Informative)
(http://oublic.org/)
Dolphin Brain on Neuroinformatica.com [neuroinformatica.com]
Once you get to the page, zoom in about ten times using the + magnifying glass icon.
The file is 135,000 pixels wide by 200,000 pixels high which would take 77.25 Gigabytes to store uncompressed. The compressed size on the server is 3.912 Gigabytes.