Loading ...
Sorry, an error occurred while loading the content.

Re: [PanoToolsNG] Using mixed FOV with 10-17mm lens in panorama

Expand Messages
  • Carel
    ... According to Bernhard it has to do with the sensor which was mentioned in this thread: http://tinyurl.com/2eso2s [quote] The basic idea arises from the
    Message 1 of 4 , Jun 2, 2007
    • 0 Attachment
      Robert Harshman wrote:
      >
      >
      > Carl, do you have the links to the references your noting?
      >
      > I'm at a lost as to why reducing the resolution could in any way
      > improve the quality if all other factors remain the same. Perhaps it
      > removes some of the complexity or imperfections of the real world?
      > Like a skin makeover in a fashion magazine?
      >
      > Regards,
      >
      > Robert
      >
      >

      According to Bernhard it has to do with the sensor which was mentioned in
      this thread:
      http://tinyurl.com/2eso2s
      [quote] "The basic idea arises from the fact that i am not content with the
      typical "pseudo-resolution" a Bayer sensor offers. I have seen in various
      situations that reducing the resolution dramatically increases
      quality."[endquote]

      The link to Michel Thoby's experiment is:
      http://tinyurl.com/2h5dzb (scroll down a bit)

      Carel
      --
      View this message in context: http://www.nabble.com/Using-mixed-FOV-with-10-17mm-lens-in-panorama-tf3858380.html#a10932495
      Sent from the PanoToolsNG mailing list archive at Nabble.com.
    • Bernhard Vogl
      ... The basic idea should be more understandable if you look how a bayer sensor actually works: Actual per-pixel color values are derived by interpolation of a
      Message 2 of 4 , Jun 2, 2007
      • 0 Attachment
        > According to Bernhard it has to do with the sensor which was mentioned in
        > this thread:
        > http://tinyurl.com/2eso2s
        > [quote] "The basic idea arises from the fact that i am not content with the
        > typical "pseudo-resolution" a Bayer sensor offers. I have seen in various
        > situations that reducing the resolution dramatically increases
        > quality."[endquote]
        >
        The basic idea should be more understandable if you look how a bayer
        sensor actually works: Actual per-pixel color values are derived by
        interpolation of a 2x2 field.
        If you reduce resolution, you can approximate to a per-pixel color
        accuracy comparable to what a Foveon sensor can deliver. See the
        following test for image quality comparison of both sensors:
        http://www.ddisoftware.com/reviews/sd9-v-bayer/
        The reality is somewhat more complex as the test may suggest, but it's a
        good aid for pointing out the basic principle (of resolution reduction).

        Best regards
        Bernhard
      Your message has been successfully submitted and would be delivered to recipients shortly.