September 25, 2010

Adobe demos refocusable images

At NVIDIA’s technology conference this week, Adobe researcher Todor Georgiev demonstrated GPU-accelerated processing of plenoptic images. As Engadget puts it, “Basically, a plenoptic lens is composed of a litany of tiny “sub-lenses,” which allow those precious photons you’re capturing to be recorded from multiple perspectives.” Plenoptic image capture could open the door to easier object matting/removal (as the scene can be segmented by depth), variable perspective after capture, and more.

This brief demo takes a little while to get going, but I still think it’s interesting enough to share.

Posted by John Nack at 6:57 AM on September 25, 2010


  • Penny Ann Dolin — 10:27 AM on September 25, 2010

    Very very interesting – (but a far better presentation of this feature/effect could be made – hope someone does it)

  • GuyE — 11:04 AM on September 25, 2010

    Holy cow

  • SBG — 11:30 AM on September 25, 2010

    How does something like this get applied practically? Buy all new lenses for your camera? Thread on some kind of filter version? How many stops do you lose?

    • Todor — 10:36 AM on October 21, 2010

      No, it’s a microlens array at the sensor. You don’t lose stops.
      It will probably come as a new camera that’s less expensive and better quality by capturing refocusable after the fact 3d.

  • skocko — 5:09 AM on September 26, 2010

    Sherman, set the WABAC machine for November 22, 2005.

    Golly, Mr. Peabody, where are we going?


  • Kevin Jay — 8:37 AM on September 27, 2010

    Apparently there is a plenoptic camera on the market.

  • imajez — 4:19 AM on September 28, 2010

    I knew about this concept, but seeing it action is something else.
    Very impressive.

Copyright © 2020 Adobe Systems Incorporated. All rights reserved.
Terms of Use | Privacy Policy and Cookies (Updated)