Slashdot Mirror

Pretty much my take too.

He may have a boat load of resolution, but until he finds a way to use it, it is largely a technical exercise.

Not more dramatic than these: http://70gigapixel.cloudapp.net/ or these http://www.dresden-26-gigapixels.com/dresden26GP but possibly done
with cheaper methods.

Still, people have been stitching images like this for a long time with equally cheap equipment, although they can't capture motion.

I have helped with similar tasks previously, where an Egyptologist had used a digital camera on a portable frame to take mosaic images of rock carvings.

http://tawbaware.com/forum2/viewtopic.php?t=1849&highlight=egypt+terje

Modern panorama software, both open source and very cheap commercial offerings, makes it possible to determine semi-automatically how multiple images are connected, while correcting for things like lens projection errors, camera location offsets and even some exposure correction.

In your case I would use a panorama head on a tripod, put the maps on top of a square grid, and then take sufficient number of images to get the final resolution you think you need.

The part of the grid that remains visible makes it easy to flatten the resulting mosaic.

I used PTAssembler on that Egyptian mosaic, highly recommended!

http://www.tawbaware.com/ptasmblr.htm

Terje

I have helped with similar tasks previously, where an Egyptologist had used a digital camera on a portable frame to take mosaic images of rock carvings.

http://tawbaware.com/forum2/viewtopic.php?t=1849&highlight=egypt+terje

Modern panorama software, both open source and very cheap commercial offerings, makes it possible to determine semi-automatically how multiple images are connected, while correcting for things like lens projection errors, camera location offsets and even some exposure correction.

In your case I would use a panorama head on a tripod, put the maps on top of a square grid, and then take sufficient number of images to get the final resolution you think you need.

The part of the grid that remains visible makes it easy to flatten the resulting mosaic.

I used PTAssembler on that Egyptian mosaic, highly recommended!

http://www.tawbaware.com/ptasmblr.htm

Terje

Ummm...call it what they like, but that's scanned film.

While not a gigapixel sensor, there is a guy that stitched together a gigapixel image from 196 digital photos, and he did this 3 years ago.

http://www.tawbaware.com/maxlyons/gigapixel.htm

Average point and shoot resolution on average 400ASA film is nowhere near 22MP, except under ideal conditions, which you won't get with point and shoot cameras in the first place. Above about 4-6MP seems to me to be a good enough compromise for anything below A4 size prints or serious art exhibitions, according to the site linked previously whose author seems to sound like he knows what he's talking about.

The big advantage of digital is that you can take multiple exposures on a tripod and blend them very easily indeed : Max Lyons has some impressive images on his site of digital panoramics using stitching, and multiple exposures blended in Photoshop to give high contrast scenes a lower contrast rendering and the results look very nice indeed.

Done that. Hell, we've even slashdotted that one before.

Check out the grand canyon in gigapixel glory

I've got to agree here. I was pretty impressed with the write up- using a computer driven servo to snap the shots... then I saw what it was they took.

I mean, really- it's a building.

Its is NO WHERE near the quality of what people like Max Lyons have done, by HAND. They wrote a script and attached a few servos.

That isn't to say what they accomplished isn't impressive- it is. But it lacks the flavour and the art of someone that's striven for a better capture method. I salute them for driving it with a servo- thats grand. Now take that setup out to some impressive scene and photograph that (Say the Tetons of Gigapixel fame) and lets see what it can do.

And if I can drive that servo with my palm pilot, plus some 26aH batteries, then I'm even more interested ;)

But until they photograph something worthy of being captured at that resolution, it'll be no different then sticking a couple of webcams out the window.

Just my opinion.

I've got to agree here. I was pretty impressed with the write up- using a computer driven servo to snap the shots... then I saw what it was they took.

I mean, really- it's a building.

Its is NO WHERE near the quality of what people like Max Lyons have done, by HAND. They wrote a script and attached a few servos.

That isn't to say what they accomplished isn't impressive- it is. But it lacks the flavour and the art of someone that's striven for a better capture method. I salute them for driving it with a servo- thats grand. Now take that setup out to some impressive scene and photograph that (Say the Tetons of Gigapixel fame) and lets see what it can do.

And if I can drive that servo with my palm pilot, plus some 26aH batteries, then I'm even more interested ;)

But until they photograph something worthy of being captured at that resolution, it'll be no different then sticking a couple of webcams out the window.

Just my opinion.

I've been looking into working with panoramic images with my digital rebel and ran across this site. There are some really beautiful panoramic photographs of Washington DC landmarks in the gallery. Max also created a gigapixel image of Bryce Canyon using 196 photographs taken with a 6 megapixel digital camera.

I've been looking into working with panoramic images with my digital rebel and ran across this site. There are some really beautiful panoramic photographs of Washington DC landmarks in the gallery. Max also created a gigapixel image of Bryce Canyon using 196 photographs taken with a 6 megapixel digital camera.

Breaking the Gigapixel Barrier
Posted by michael on Tuesday December 02, @05:06PM
from the sweetness dept.
megas writes "Max Lyons has just posted on his site what seems to be the first 1 Gigapixel picture, created from 196 separate photographs taken with a 6 megapixel digital camera, and then stitched together into one seamless composite. According to Max, he has 'been unable to find any record of a higher resolution photographic (i.e. non-scientific) digital image that has been created without resizing a smaller, lower resolution image or using an interpolated image.'"

Breaking the Gigapixel Barrier
Posted by michael on Tuesday December 02, @05:06PM
from the sweetness dept.
megas writes "Max Lyons has just posted on his site what seems to be the first 1 Gigapixel picture, created from 196 separate photographs taken with a 6 megapixel digital camera, and then stitched together into one seamless composite. According to Max, he has 'been unable to find any record of a higher resolution photographic (i.e. non-scientific) digital image that has been created without resizing a smaller, lower resolution image or using an interpolated image.'"

I guess this guy is going to be somewhat disappointed when he hears about this.

Gigapixel photography using several camera's is not nearly new.

As noted, the details on this are sketchy, but I HIGHLY doubt it's a 70 megapixel camera - just a rotater/stitcher on top of a run-of-the-mill digicam platform. What they probably make easy for you is keeping the exposure exact, the shooting location the same, and stitching it all togather for you.

You too can do this (with a little effort) - for example, here is a Longs Peak Panorama generated by stitching togather a dozen or so pictures from a Canon G5 consumer digicam. It does look a bit wierd seeing a rotated view on a flat surface though.

BTW, along the lines of that Grand Canyon shot, Page 3 of the original article also showed a stitched panorama of Pike Peak, Colorado - if you are interested, here is series of pictures of Longs Peak Colorado which includes a Longs Peak Panorama similar to the one in the original article - this is yet another example where digital makes stuff like this easier.

Besides ... you could easily take GIGApixel images with your existing camera, using the stiching technique outlined on Max Lyons' website.

I have yet to try this with my Canon G3 ... no time, no subject matter ... but if the results are anything like Max's, I'll be impressed.

Okay, this is all very impressive and all.

But I still don't quite get why the guy didn't just take a whole bunch of normal pictures and warp/stich them together. This technology has been around for a while, and it can be automated so his throughput could be much higher than 5-8 per year. The article mentions "seaming multiple photos" together, but doesn't say why that wasn't an option. I wonder if it's a matter of the photographer's technical expertise.

For proof that stitching multiple photos works well, especially for outdoor nature scenes, check out this web site that describes the work of a guy who took many pictures of Bryce Canyon National Park in Utah. After stitching together 196 pictures, he ended up with one digital image that measures 40,784 x 26,800, with 1.09 billion pixels. In a raw format at 24 bit, that's way over the 2.6 gigabytes from the parent article.

One gigapixel is one gigapixel. Perhaps you meant to ask about the size on disk? The Max Lyons picture (recently exhibited) chomps up 2,068,654,055 bytes of diskspace somewhere. He achieves a LOT with affordable means, and is a also a wonderful photographer. I never tire of that site.

Here.
It was /.'ed back in December....where u take a regular high-megapixel camera, and a specialized tripod and take many pictures and then use software to stitch 'em together.

All the CCDs I've dealt with have significant thermal noise, so exposures must be kept short unless the device is kept very cold. Since the point of a pinhole camera is to have the smallest aperture possible, wouldn't you need fairly long exposures to get enough light onto the sensor?

There are tools like Pixelzap that do darkfield subtraction which can help with the pixels that are always noisy, but since thermal noise is random, there's always some to contend with. How would a digital pinhole camera deal with this?

It's called hugin, and is available here. It even outputs layered files which capitalize on capabilities recently added to the Gimp 2.0!

Take a look! You might also like to see the worlds first gigapixel image created using the techniques hugin employs here.

Not quite that, but this guy seems to have taken the concept of stitching low-res high-zoom photo's together to get high-res to a new level.

C:\>netstat -a | findstr LISTENING >file.txt
C:\>wc -l file.txt
file.txt: Lines: 12
C:\>ver
Microsoft Windows XP [Version 5.1.2600]
C:\>

wc binary link here

"This is a 360-degree view," said Michael Malin, a Spirit science team member and president of Malin Space Science Systems. "This is multiple thousands of pixels high and many tens of thousands of pixels wide. In fact, being shown here is something that we assembled at one half the resolution and one quarter the number of pixels because the sheer number of pixels was slowing down our machine in getting it ready for you today. We felt it as better to get it out and we'll show you a little later on, maybe in the next couple of days, a better view of it. This view in itself will jam any download you ever tried to perform."

I guess you haven't seen this yet, then, have you?

Yes, I realize it's not very practical for sports or photojournalism, but this is only going to get better and cheaper. Everyone who's bought a decent digital camera will tell you the same thing: for 90% of my work, digital does the same thing as film, only it's a shitload cheaper, a shitload easier, and offers some fantastic additional benefits. Think of media storage for instance -- storing slides or negs is a bitch, whether you're a pro dealing with cataloging thousands of images for business, or you're an amateur with a dozen shoe-boxes of holiday and travel shots. Digital makes this so easy it hurts.

Now, you can certainly argue the merits of film technology not requiring as much continued investment, but the fact is, the pro-sumer line of cameras that are out now rival film in all characteristics save one: tonal range. The room for new technological growth is still there, but at this point the 35mm evolution to digital is complete.

People that argue about resolution are missing the bigger picture: if I want to do anything with an image, whether digital or analog, the first thing I'm going to do is get it into my computer. That's easier when the format I'm shooting in is already digital. Also, if I'm scanning a slide, even on a *nice* scanner, you're not going to see any improvement over the 5 meg files I get out of my digital body. What you *will* see is lots of dust, which means a few hours Photoshopping. Most of the time, a sub $20k scanner's extra pixels are just interpolation, anyway. There's plenty of software that can do that with low-res images already.

In terms of maturity -- have you seen the long-exposure capabilities of Canon's digital line? Holy-freakin-shit! Even an EOS D60, which is now outdated, can produce 4-minute exposures with no noise. Nothing. Turn the night into day.

Then there's the added benefits for learning photographers. If you want to get good, you shoot your ass off. For the first couple of years, you toss out 35/36 shots. As you get better, you'll slowly lower that, but the fact is, developing that much film is expensive. And as a learning tool, if I'm going to figure out that a blown shot at f/8 would have been perfect at f/11, I need to know right after I've taken the shot. Not a week later when I finally get my film back. And that's only useful when I've recorded the exposure for every shot. Have you ever tried this? After a single roll you'll never want to do it again.

With digital, you get instant feedback as to what you're technically doing right or wrong. Hell, nice pro-sumer digitals offer color histograms of your shots. I can confidently say that with the right teacher, a digital camera will allow an amateur to develop the technical skills of a pro in under a year (now, the artistic skills may never come, but that's another issue entirely).

When you get into bigger boxes (8x10's and the like) you're talking about thousands of dollars of investment for good glass and equipment (and good luck with your processing costs -- you can always buy an enlarger!). Medium format equipment can run you several times more if you want the "35mm experience" like the fancy Mamiya 645's. Frankly, I don't see any advantage to traditional film unless you: 1) Already know what you're doing, and 2) Are currently making a living off of it. And even then I'd recommend it, unless you 3) Have already spent a huge chunk on medium or large-format, and are too unsophisticated to figure out how to "work the eBay".

Clearely this must beat the giga pixel stitch which I remember from not long ago. OK arguably not the same thing, but I had that in mind when I read about the size of the images.... The result is a massive mosaic of digital images 85 gigabytes big. I mean, Geez :)

I know - it is painful to admit that your pile of very expensive toys is obsolete.

Good digital back already has more information then a medium format film (like scanning backs). Combined with far superiour capabilities of a digital darkroom image quality is already leagues ahead.

Sell your toys while you still can. Nobody will give a dime for them few years from now.

If you just scale typical sensor used in smaller cameras to the new 4/3 format, you get 30Mp+. Best medium format lenses start choking at this level. And Photoshop violently rapes darkroom processing capabilities in the ass.

Film is pretty much dead.

Would a 100px tall strip of the image make you happy?

http://www.tawbaware.com/maxlyons/gigapixel_strip. jpg

It's only a cropping, though.. 40784 x 100 pixels. /. away :)

• And when you're a professional (or semi-professional), which is what this product is aimed at, you're probably not shooting .jpg anymore.

Have you checked Tabaware? The author of Pixelzap and others, offers a ton of info on post-processing digital images to do neat tricks, including perspective correction for panorama assembly, etc.

The biggest advantage that film does have - it will continue to enjoy for some time to come - is dynamic range.

I don't buy that: how many stops of dynamic range does slide film have? Not to mention that with digital you can do really interesting things like taking multiple exposures of the same scene and combining them for some really impossible-to-repro-with-film results.

this out for example (check these 5 composite images)

let alone that in an Ansel Adams or Weston print.

you're talking apples & oranges here: those prints have been hand-developed, dodged, burned and so on, you can do the same (increasing apparent film dynamic range) in photoshop and print the results if you so choose.

A master with the film camera, will probably produce masterful digital pics very quickly, if you look at the pictures of the week on photo.net, you'll probably see some stunning digital shots, which would have been just as stunning in shots on film, and sometimes even more so due to the possibilities inherent in a digital imaging workflow.