In the future, we might be able to roam around a 3D virtual representation of our world, where everything you see was automatically generated from photographs taken at the real locations.

Vision researchers at the University of Washington and Cornell University have been working on turning photographs of things in the real world into 3-dimensional representations. This research could eventually turn snapshots into virtual buildings, neighborhoods, and possibly cities.

PhotoCity is a new online game created by researchers that aims to harness the power of crowdsourcing in order to obtain the photographs needed for reconstructing these locations. Here’s a short 1 minute introduction of how the game works:

In addition to playing with a 2 or 3 megapixel camera, they’ve also released an iPhone app:

We’re pretty darn excited to see how photography will play a role in the technologies we’ll be using in the future.

Thanks for the tip, @eugenephoto

P.S. If you haven’t seen the TED 2007 demo of Microsoft’s Sea Dragon and Photosynth technologies, you’re in for a treat.

Nikon recently conducted a survey called “Picture Yourself”, and released some interesting findings today, a lot of which reveals things about the way people view themselves. The survey was conducted using random telephone dialing, and sampled 1000 Americans 18 and older.

If only we had thought of some of these questions for our weekly polls… Maybe we’ll use a few in the future.

Anyhow, here were some interesting findings:

• 25% would choose to retake a family photo if they could turn back time
• 92% of adults had at least one photograph taken of them so far this year
• 79% believe they look better in person than in photographs (do you?)
  
• 26% chose weight as the feature they dislike the most in photos
• Men have an average of 73 photos taken of them, while women have only 38

Are there any other questions you wish Nikon had included in the survey? Let us know, and we’ll conduct “research” with PetaPixel polls!

(via The Imaging Resource)

Image credit: Mirrored self-misidentification by eqqman

Two days ago, Sankei News in Japan reported that a Nagoya Institute of Technology team led by Professor Yojiro Ishino was certified by Guinness World Records as having built the “Camera With The Most Lenses”. The camera boasts a staggering 158 separate lenses.

The setup reminds me of the camera array used to shoot the “bullet time” scenes in The Matrix, except rather than having a large number of individual cameras, this setup has a single “camera” with a large number of lenses.

The record-breaking camera took six months to build, and is meant to capture the movement of a flame simultaneously from multiple directions. Each lens cost only 200 yen, which is about $2.26. Thus, the whole collection of lenses cost only about $360.

Kudos if anyone can send us photographs taken with this camera!

(via CrunchGear)

Update: Here’s an image taken with the camera, and some additional photographs of the camera itself. Kudos to Durf for finding this!

The web is abuzz over a project over at Stanford that aims to revolutionize how we think about photography by building an open source camera (dubbed Frankencamera).

That’s right…

Open. Source. Camera.

While you try to wrap your mind around this new paradigm, I’ll point out of a few of the important aspects of the project and throw in some of my thoughts on it.

Linux, Firefox, and now Frankencamera

The established order of things up to this point has been for behemoth camera corporations (i.e. Canon, Nikon, etc…) to sell consumers (i.e. you and me) hardware and software that they spend years and billions of dollars developing and tweaking. The same was true of operating systems and browsers before open source projects like Linux and Firefox crashed the party.

If this research group at Stanford successfully releases an open source platform for imaging, a whole new world of opportunity opens up for photographers and developers alike. Instead of attempting to have features added to future cameras by making noise and requesting them, we would be able to take matters into our own hands, building hardware or developing software to suit our needs.

Advanced In-Camera “Post”-Processing

Imagine if you could program your DSLR with some common Photoshop actions that you always run when post-processing your images, so the photos come out the camera with your edits already applied.

Even more advanced post-processing techniques could be moved into the camera, providing photographers with features that the large camera makers would never add to their DSLRs, since they prefer sticking to the fundamentals and leaving post-processing up to the photographer. For example, a photographer could choose to have his camera automatically bracket, merge, and tone map, allowing him to download HDR photographs directly from his camera.

Camera Apps

The team behind the Frankencamera also envisions a future where photographers can download applications onto their cameras, just like apps can be downloaded to the iPhone from the App Store. Wifi on your camera? Directly uploading photographs to Flickr? Different photo styles and camera effects? The possibilities are endless, and it would definitely be interesting to see what applications developers would come up with.

As Apple’s App Store has shown, it definitely pays to put application development in the hands of individuals rather than keep it behind closed doors with your relatively small group of developers and engineers.

Hardware

How would an open source software platform change the game in terms of hardware? The Frankencamera is currently being developed with a hodgepodge of parts — everything from Nokia cameraphone sensors to Canon lenses. If an open source camera gained any significant piece of the camera market pie, then third party lens manufacturers such as Tamron and Sigma would no doubt join in on the fun.

I’ve read elsewhere that third-party lens makers are forced to reverse engineer the mounting and focusing systems of camera makers such as Canon and Nikon. This would be completely unnecessary for an open source camera, and the third-party companies would even be able to contribute towards the software side to improve the functionality of their lenses.

A critical piece of the puzzle, however, is the issue of sensors. I’m sure the bulk of the billions spent on R&D has to do with sensor technology, and pretty much no one can compete with the larger companies on this front. No matter how popular an open source camera might be, adopters will likely have to take a hit on sensor quality unless one of the big players decides to contribute their sensors.

Final Thoughts

Personally, I think this is a great idea and really hope the research group succeeds in getting something off the ground and into our hands. I only wish it were a project being done over here at UC Berkeley, though I do know there’s some pretty interesting work being done related to camera sensors and bokeh rendering.

To learn more about the project, you can read the Stanford news article, or watch this YouTube video put out by Stanford.