Being an amateur photographer, and I do emphasize the word amateur, I’ve been slowly collecting my ordnance of photographic equipment – a basic, run-of-the-mill digital SLR body (which I’ll have to eventually upgrade), a couple of lenses, a tripod, a flash, and extra memory. Although the increased flexibility and control of owning an SLR is definitely appreciated, all of this equipment did come at a cost.  I also don’t have the option of manipulating my photos unless I pull out my notebook computer, import my photos and then open Photoshop (or Photoshop Lightroom for those of you who don’t want to or can’t afford to fork out the extra money).

The rise in mobile computational devices such as smartphones and the opening of SDKs for operating systems such as Android, the iPhone, and Symbian have resulted in the proliferation of applications that have enabled users the ability to edit photos without having to purchasing expensive third party applications. Some of these are even free. For example, instead of working in Photoshop to create panoramic photos, I simply purchased a fantastic little app called Panorama for my iPhone and it uses an algorithm to seamlessly patch photos together.

However, the compromise in using a mobile device to take photos is that the quality of the camera is nowhere near that of using a DSLR. Beyond ‘face detection’ and ‘red eye reduction’, how will camera companies continue to further the computational development of their products?

Stanford Computer Science researchers have gone beyond this question and are taking matters into their own hands. Marc Levoy, professor of Computer Science and Electrical Engineering, and his students are currently developing what they call ‘Frankencamera’, an open-source and fully programmable camera that allows computational photography researchers and enthusiasts to develop and test new ideas and applications. They believe that the camera no longer needs to be limited by the features that a closed and proprietary camera manufacturer deems fit to supply. All the features of the Stanford camera – focus, exposure, shutter speed, flash, etc. – are at the command of software that can be created by inspired programmers anywhere.

I think Levoy phrases it best when he states:

For the last ten years there’s been a megapixel war. That’s finally winding down and the asian camera manufacturers can no longer compete based on the number of megapixels so now they’re beginning to compete on one other feature. We’d like them to offer lots and lots of features that the research community is currently working on and offer those all at one; not merely when they want to compete with one another.

With support from Nokia, Adobe Systems, Kodak, and Hewlett-Packard, Levoy and his students have put together the open-source camera from a number of difference parts. The motherboard is a Texas Instruments ‘system on a chip’ running Linux with image and general processors and a small LCD screen. The imaging chip is from a Nokia N95 cell phone. They are using standard Canon lenses, but they are combined with actuators to give the camera fine-tuned software control. Finally, the body is custom made at Stanford.

Within about a year, after the camera is developed to his satisfaction, Levoy hopes to have to have the funding and the arrangements in place for an outside manufacturer to produce them in quantity, ideally for less than $1,000. Levoy would then provide them at cost to colleagues and their students at other universities.

Although far from distribution to the public, this is an incredible advancement in computational photography. One can only hope that camera manufacturers will catch word of this and will adopt their philosophy, accelerating the process with their use of capital and technology.