One Billion Pixels, But Are They Any Good?

Max Lyons has created some terrific panoramic images, each comprised of a matrix containing several hundred digital photographs. The individual shots are assembled, processed, and rerendered using software to correctly align and recolour each image to give the illusion of a single photo. The end result is a photograph with just over 1.09 billion pixels - the first amateur photograph to break the gigapixel barrier.

The images have so much detail, that it is possible to discern more from the images than would be possible with the naked eye, even when viewed with binoculars:

Assuming one pixel per arcminute, an image with dimensions of 3780 x 2485 would suffice to capture the amount of detail that the naked eye could resolve. This image has more than 100 times this detail. Looking at the full sized digital image, one is able to see things that might have been difficult or impossible to spot, even when using binoculars.

This imaging technique lends itself well to creating extremely high-resolution head-mounted simulations. It is also cost-effective (you only need one camera), and scalable to meet resolution requirements (take more photographs).

An interesting question arises: namely, is the resulting image (that is, the output to a high-resolution head-mounted, panoramic display that responds to head movements) distinguishable from the real thing? Ignoring, for a moment, the absence of other sensory correspondence (and the obvious presence of the head-mounted unit), I think I'd have difficulty making the distinction. What we have here is actually a more accurate representation of the scene than perceivable reality. (Though the theoretical capabilities of the human eye are often underestimated).

Of course, imaging techniques have a long way to go before we start seeing viable ways to actually capture motion at this level of detail. Until then, all we're left with are some very detailed billboards and somewhat pretty postcards.