On Jun 22, 2011, at 6:59 AM, Godfrey DiGiorgi wrote: > On Tue, Jun 21, 2011 at 3:20 PM, Larry Colen <[email protected]> wrote: >> Redundant Array of Inexpensive DSLRs.... >> >> On my drive home from San Luis Obispo today, I had the crazy idea that for >> the price of a medium format camera, you could make a bracket that would >> securely hold four (or more) relatively inexpensive DSLRs, with identical(*) >> decent lenses, trigger them simultaneously, and use software to combine the >> images into one extremely high resolution, or high dynamic range, image. >> This would have the advantage of all of the shots being taken at the same >> time, and for landscape photos, the parallax difference would likely be >> negligible. >> >> I'm curious of any one knows of any work along these lines... stitching >> images from multiple cameras, taken at the same time, rather than a single >> camera taken over a longer span of time. >> >> >> * By identical I mean the same make and model, individual variation is >> inevitable. > > Ach, getting all those cameras to have a coincident focus setting, > exposure setting, response curve, etc ... Getting the shutters to > respond in sync ... you're going to slow down the picture making > process by a bunch. Not a bad thing, but kind of a specialist > endeavor.
It's definitely a specialist endeavor. Four K-5s with FA31s, or even used K20s or rebels with decent primes would be cheaper than a 645D with equivalent glass, but not by much. Though it's the sort of thing that could be fun to do on a group outing, using several people's cameras. If part of the goal is to expand dynamic range, then coincident exposure setting isn't an issue. > > Even simple stereo cameras with two lenses/shutter systems (Stereo > Realist, Stereo Rolleis from the 1920s and 1930s) show how important > the distance between the optical centers of the lens systems can be. > We can do more with processing nowadays. Image processing has come a fair way in the past 80 or so years. > > But it is an interesting thought exercise. I do remember seeing a > video of a setup designed for some kind of capture effort that had a > few dozen Canon 5Ds with identical lenses linked together for > coordinated capture of a formula 1 car. The din of making the exposure > was breathtaking. Now that you mention it, I remember seeing that video. Can't seem to find it. Combining responses to other posts, Ctein's suggestion is along very similar lines to my idea. His would be cheaper for mass market, mine is cheaper for onesy twosy, especially if I could borrow cameras. The trick is getting software that would do a good job of combining the photos. I'm familiar with the plenoptic concepts, and I think that thinking along the lines of that, or along the lines of Ctein's idea was where I started, but I realized that it would be easier to build a bracket to hold several cameras than it would to build custom photographic hardware. Apart from mirrorless camera sensors not yet being quite as favorable on the price/performance curve as a used DSLR, they could probably be packed tighter for less parallax. I'm sure that someone more proficient at math than I can figure out minimum shooting distance for each focal length at a particular pixel density were the parallax difference is less than a pixel. The problem with multiple exposures separated in time, rather than space is that even on a relatively calm day, things move. You couldn't use that technique for surf, rivers, or even trees in any sort of a wind. My control system would be a bunch of electronic cable releases wired in parallel to a single switch. -- Larry Colen [email protected] sent from i4est -- PDML Pentax-Discuss Mail List [email protected] http://pdml.net/mailman/listinfo/pdml_pdml.net to UNSUBSCRIBE from the PDML, please visit the link directly above and follow the directions.
