Thalassa
Jun 22, 11, 11:30 am
A California-based startup called Lytro (http://www.lytro.com) has introduced a pretty exciting new imaging technology that allows images to be e.g. refocused after shooting.
The company is promising consumer products during 2011 so we should see the first results pretty soon.
The science behind the product seems legit, as the founder and CEO of the company has won several prices (including the ACM award (http://www.acm.org/press-room/news-releases-2007/internal/) for the best dissertation of the year). The dissertation is available through the company website, but for those who do not wish to read a 160 page treatise, there is a fairly approachable scientific paper available here (http://graphics.stanford.edu/papers/lfcamera/lfcamera-150dpi.pdf).
The technology is in its infancy and cannot do miracles (there are limits to the extra DOF it can capture and resolution is sacrificed), but this is potentially one of the big breakthroughs in digital imaging. (And, yes, I do know there have been scientific papers about aspects of lightfield photography and plenoptic lenses since the early 90's, so there is no need to point that out. Most of the earlier research is simply not very practical.)
The demo images on the company website with shiftable focal plains suggest a pretty cool application: combine a plenoptic image (or even a conventional image with focus-stacking information) with an eye-movement sensor and you can create pretty fantastic "live" images that show different focal points depending on where one looks.
Cheers,
T.
The company is promising consumer products during 2011 so we should see the first results pretty soon.
The science behind the product seems legit, as the founder and CEO of the company has won several prices (including the ACM award (http://www.acm.org/press-room/news-releases-2007/internal/) for the best dissertation of the year). The dissertation is available through the company website, but for those who do not wish to read a 160 page treatise, there is a fairly approachable scientific paper available here (http://graphics.stanford.edu/papers/lfcamera/lfcamera-150dpi.pdf).
The technology is in its infancy and cannot do miracles (there are limits to the extra DOF it can capture and resolution is sacrificed), but this is potentially one of the big breakthroughs in digital imaging. (And, yes, I do know there have been scientific papers about aspects of lightfield photography and plenoptic lenses since the early 90's, so there is no need to point that out. Most of the earlier research is simply not very practical.)
The demo images on the company website with shiftable focal plains suggest a pretty cool application: combine a plenoptic image (or even a conventional image with focus-stacking information) with an eye-movement sensor and you can create pretty fantastic "live" images that show different focal points depending on where one looks.
Cheers,
T.