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

I need a test image to show the difference between Lanczos and BiCubic iterpolation...

Expand Messages
  • Ken Warner
    I ve written my own Lanczos and bicubic interpolators. When I apply them to the typical equirectangular pano, I really don t see much difference. I m trying
    Message 1 of 8 , Apr 5, 2009
    • 0 Attachment
      I've written my own Lanczos and bicubic interpolators. When I apply
      them to the typical equirectangular pano, I really don't see much difference.

      I'm trying to talk myself into believing that the Lanczos interpolator is
      better than the bicubic interpolator but I can't. I just can't see the
      difference. In fact I can barely see the difference between bilinear
      and bicubic interpolation. Bilinear is only slightly softer.

      So what I need is a test image that I can use to show the difference
      to myself. Dr. Dersch used some images with concentric circles --

      http://home.no.net/dmaurer/~dersch/interpolator/interpolator.html

      Where does one get a test image like that?

      Or does anyone know another image that will show defects and clarity
      of interpolation?
    • Gerald Lodron
      I also made a few test images (also that one you which is used by Dr. Dersch) and loaded them on fotolia: http://de.fotolia.com/p/200943235/partner/200943235
      Message 2 of 8 , Apr 5, 2009
      • 0 Attachment
        I also made a few test images (also that one you which is used by Dr. Dersch) and loaded them on fotolia:
        http://de.fotolia.com/p/200943235/partner/200943235

        Its a simple 2d mathematical function of a sine with increasing frequency.

        Best regards



        ________________________________
        From: Ken Warner <kwarner000@...>
        To: PanoToolsNG@yahoogroups..com
        Sent: Monday, April 6, 2009 2:04:13 AM
        Subject: [PanoToolsNG] I need a test image to show the difference between Lanczos and BiCubic iterpolation...


        I've written my own Lanczos and bicubic interpolators. When I apply
        them to the typical equirectangular pano, I really don't see much difference.

        I'm trying to talk myself into believing that the Lanczos interpolator is
        better than the bicubic interpolator but I can't. I just can't see the
        difference. In fact I can barely see the difference between bilinear
        and bicubic interpolation. Bilinear is only slightly softer.

        So what I need is a test image that I can use to show the difference
        to myself. Dr. Dersch used some images with concentric circles --

        http://home. no.net/dmaurer/ ~dersch/interpol ator/interpolato r.html

        Where does one get a test image like that?

        Or does anyone know another image that will show defects and clarity
        of interpolation?


        __._,_..___
        Messages in this topic (1) Reply (via web post) | Start a new topic
        Messages | Files | Photos | Links | Database | Polls | Members | Calendar
        --

        Change settings via the Web (Yahoo! ID required)
        Change settings via email: Switch delivery to Daily Digest | Switch format to Traditional
        Visit Your Group | Yahoo! Groups Terms of Use | Unsubscribe
        Recent Activity
        *  3
        New MembersVisit Your Group
        Give Back
        Yahoo! for Good
        Get inspired
        by a good cause.
        Y! Toolbar
        Get it Free!
        easy 1-click access
        to your groups.
        Yahoo! Groups
        Start a group
        in 3 easy steps.
        Connect with others.
        .






        [Non-text portions of this message have been removed]
      • Fulvio Senore
        You can see some example in this page: http://www.fsoft.it/panorama/FSPViewer.htm there are bilinear and some Lanczos ones, no bicubic, sorry. Bicubic is a
        Message 3 of 8 , Apr 5, 2009
        • 0 Attachment
          You can see some example in this page:

          http://www.fsoft.it/panorama/FSPViewer.htm

          there are bilinear and some Lanczos ones, no bicubic, sorry.

          Bicubic is a rather generic term, but some time ago I looked at the
          Lanczos and bicubic implementations in PTGui and I believe that bicubic
          normal is better.
          Lanczos has a slightly sharpening effect that can cause ringing (echoes
          of sharp edges, like in overcompressed jpegs).
          For this reason IMHO Lanczos is better for viewers where a sharp image
          is more pleasant, but bicubic is better for file creation, where the
          lanczos oversharpening might not be desired.

          About the interpolators test in the Dersch page, I think that it is not
          representative of real usage. Nobody applies so many interpolations to
          the same image.
          That test shows that a large windowed sinc kernel works very well, but
          this is not always true. Just try to interpolate a screenshot with sharp
          edges and you will see large ringing effects.

          Fulvio Senore


          Ken Warner ha scritto:
          > I've written my own Lanczos and bicubic interpolators. When I apply
          > them to the typical equirectangular pano, I really don't see much difference.
          >
          > I'm trying to talk myself into believing that the Lanczos interpolator is
          > better than the bicubic interpolator but I can't. I just can't see the
          > difference. In fact I can barely see the difference between bilinear
          > and bicubic interpolation. Bilinear is only slightly softer.
          >
          > So what I need is a test image that I can use to show the difference
          > to myself. Dr. Dersch used some images with concentric circles --
          >
          > http://home.no.net/dmaurer/~dersch/interpolator/interpolator.html
          >
          > Where does one get a test image like that?
          >
          > Or does anyone know another image that will show defects and clarity
          > of interpolation?
          >
          >
          >
          > ------------------------------------
          >
        • Ken Warner
          Thanks... but not for 20 euros...
          Message 4 of 8 , Apr 6, 2009
          • 0 Attachment
            Thanks... but not for 20 euros...

            Gerald Lodron wrote:
            > I also made a few test images (also that one you which is used by Dr. Dersch) and loaded them on fotolia:
            > http://de.fotolia.com/p/200943235/partner/200943235
            >
            > Its a simple 2d mathematical function of a sine with increasing frequency.
            >
            > Best regards
            >
            >
            >
            > ________________________________
            > From: Ken Warner <kwarner000@...>
            > To: PanoToolsNG@yahoogroups..com
            > Sent: Monday, April 6, 2009 2:04:13 AM
            > Subject: [PanoToolsNG] I need a test image to show the difference between Lanczos and BiCubic iterpolation...
            >
            >
            > I've written my own Lanczos and bicubic interpolators. When I apply
            > them to the typical equirectangular pano, I really don't see much difference.
            >
            > I'm trying to talk myself into believing that the Lanczos interpolator is
            > better than the bicubic interpolator but I can't. I just can't see the
            > difference. In fact I can barely see the difference between bilinear
            > and bicubic interpolation. Bilinear is only slightly softer.
            >
            > So what I need is a test image that I can use to show the difference
            > to myself. Dr. Dersch used some images with concentric circles --
            >
            > http://home. no.net/dmaurer/ ~dersch/interpol ator/interpolato r.html
            >
            > Where does one get a test image like that?
            >
            > Or does anyone know another image that will show defects and clarity
            > of interpolation?
            >
            >
            > __._,_..___
            > Messages in this topic (1) Reply (via web post) | Start a new topic
            > Messages | Files | Photos | Links | Database | Polls | Members | Calendar
          • Jim Watters
            I did my own test here http://photocreations.ca/interpolator/index.html I used the test image from Ken Turkowski s Zone Plate
            Message 5 of 8 , Apr 6, 2009
            • 0 Attachment
              I did my own test here
              http://photocreations.ca/interpolator/index.html
              I used the test image from Ken Turkowski's Zone Plate
              http://www.worldserver.com/turk/opensource/index.html#ZonePlate


              Fulvio Senore wrote:
              > You can see some example in this page:
              >
              > http://www.fsoft.it/panorama/FSPViewer.htm
              >
              > there are bilinear and some Lanczos ones, no bicubic, sorry.
              >
              > Bicubic is a rather generic term, but some time ago I looked at the
              > Lanczos and bicubic implementations in PTGui and I believe that bicubic
              > normal is better.
              > Lanczos has a slightly sharpening effect that can cause ringing (echoes
              > of sharp edges, like in overcompressed jpegs).
              > For this reason IMHO Lanczos is better for viewers where a sharp image
              > is more pleasant, but bicubic is better for file creation, where the
              > lanczos oversharpening might not be desired.
              >
              > About the interpolators test in the Dersch page, I think that it is not
              > representative of real usage. Nobody applies so many interpolations to
              > the same image.
              > That test shows that a large windowed sinc kernel works very well, but
              > this is not always true. Just try to interpolate a screenshot with sharp
              > edges and you will see large ringing effects.
              >
              Correct, not representative of real usage but by doing the test multiple
              time you are multiplying the error so it is more visible. But
              transforms is not out.

              When creating a panorama generally an image is transformed 5 or less
              times.
              Once to correct image from Chromatic Aberrations, light falloff, etc...
              Second to transform from original into pan.
              Third to transform patch area to rectangular.
              Forth to transform patch back into pan.
              Fifth to create cubes faces from pan.

              > Fulvio Senore
              >
              >
              > Ken Warner ha scritto:
              >
              >> I've written my own Lanczos and bicubic interpolators. When I apply
              >> them to the typical equirectangular pano, I really don't see much difference.
              >>
              >> I'm trying to talk myself into believing that the Lanczos interpolator is
              >> better than the bicubic interpolator but I can't. I just can't see the
              >> difference. In fact I can barely see the difference between bilinear
              >> and bicubic interpolation. Bilinear is only slightly softer.
              >>
              >> So what I need is a test image that I can use to show the difference
              >> to myself. Dr. Dersch used some images with concentric circles --
              >>
              >> http://home.no.net/dmaurer/~dersch/interpolator/interpolator.html
              >>
              >> Where does one get a test image like that?
              >>
              >> Or does anyone know another image that will show defects and clarity
              >> of interpolation?
              >>
              >>
            • Ken Warner
              Thanks -- I turned it into a 2000x1000 equirectangular. I still can t see much (if any) significant difference between the three different interpolators --
              Message 6 of 8 , Apr 6, 2009
              • 0 Attachment
                Thanks -- I turned it into a 2000x1000 equirectangular.

                I still can't see much (if any) significant difference between
                the three different interpolators -- bilnear, bicubic, Lanczos
                at web resolutions.

                Here's the path to the image if you want to put it in your own viewer.

                http://pancyl.com/images/ZonePlatePanorama.jpg

                If you want to bore yourself -- the interpolators are in my Java viewer.

                http://pancyl.com/ZonePlatePanorama.htm

                If you press the 'i' key, the viewer rotates between bilinear,
                bicubic and Lanczos interpolators. When the dot turns red, it's
                done interpolating. The intermediate stage is nearest neighbor and
                that is indicated with the orange dot.

                Fullscreen: F1 enter / ESC exit.


                Jim Watters wrote:
                > I did my own test here
                > http://photocreations.ca/interpolator/index.html
                > I used the test image from Ken Turkowski's Zone Plate
                > http://www.worldserver.com/turk/opensource/index.html#ZonePlate
                >
                >
                > Fulvio Senore wrote:
                >
                >>You can see some example in this page:
                >>
                >>http://www.fsoft.it/panorama/FSPViewer.htm
                >>
                >>there are bilinear and some Lanczos ones, no bicubic, sorry.
                >>
                >>Bicubic is a rather generic term, but some time ago I looked at the
                >>Lanczos and bicubic implementations in PTGui and I believe that bicubic
                >>normal is better.
                >>Lanczos has a slightly sharpening effect that can cause ringing (echoes
                >>of sharp edges, like in overcompressed jpegs).
                >>For this reason IMHO Lanczos is better for viewers where a sharp image
                >>is more pleasant, but bicubic is better for file creation, where the
                >>lanczos oversharpening might not be desired.
                >>
                >>About the interpolators test in the Dersch page, I think that it is not
                >>representative of real usage. Nobody applies so many interpolations to
                >>the same image.
                >>That test shows that a large windowed sinc kernel works very well, but
                >>this is not always true. Just try to interpolate a screenshot with sharp
                >>edges and you will see large ringing effects.
                >>
                >
                > Correct, not representative of real usage but by doing the test multiple
                > time you are multiplying the error so it is more visible. But
                > transforms is not out.
                >
                > When creating a panorama generally an image is transformed 5 or less
                > times.
                > Once to correct image from Chromatic Aberrations, light falloff, etc...
                > Second to transform from original into pan.
                > Third to transform patch area to rectangular.
                > Forth to transform patch back into pan.
                > Fifth to create cubes faces from pan.
                >
                >
                >>Fulvio Senore
                >>
                >>
                >>Ken Warner ha scritto:
                >>
                >>
                >>>I've written my own Lanczos and bicubic interpolators. When I apply
                >>>them to the typical equirectangular pano, I really don't see much difference.
                >>>
                >>>I'm trying to talk myself into believing that the Lanczos interpolator is
                >>>better than the bicubic interpolator but I can't. I just can't see the
                >>>difference. In fact I can barely see the difference between bilinear
                >>>and bicubic interpolation. Bilinear is only slightly softer.
                >>>
                >>>So what I need is a test image that I can use to show the difference
                >>>to myself. Dr. Dersch used some images with concentric circles --
                >>>
                >>>http://home.no.net/dmaurer/~dersch/interpolator/interpolator.html
                >>>
                >>>Where does one get a test image like that?
                >>>
                >>>Or does anyone know another image that will show defects and clarity
                >>>of interpolation?
                >>>
                >>>
                >
                >
                >
                >
              • Gerald Lodron
                The price depends on the ressolution and starts at 1 euro and goes to 10 euros for the full image.: 346 x 346 (0,1 MP), 12,2 cm x 12,2 cm @72 dpi 1euro 693 x
                Message 7 of 8 , Apr 6, 2009
                • 0 Attachment
                  The price depends on the ressolution and starts at 1 euro and goes to 10 euros for the full image.:
                  346 x 346 (0,1 MP), 12,2 cm x 12,2 cm @72 dpi 1euro
                  693 x 693 (0,5 MP), 24,4 cm x 24,4 cm @72 dpi 2euro
                  etc

                  20 euros are for the enhanced licencence, thats when you want to make money with it (image on a product etc.).


                  ________________________________
                  From: Ken Warner <kwarner000@...>
                  To: PanoToolsNG@yahoogroups.com
                  Sent: Monday, April 6, 2009 12:46:04 PM
                  Subject: Re: [PanoToolsNG] I need a test image to show the difference between Lanczos and BiCubic iterpolation...


                  Thanks... but not for 20 euros...

                  Gerald Lodron wrote:
                  > I also made a few test images (also that one you which is used by Dr.. Dersch) and loaded them on fotolia:
                  > http://de.fotolia. com/p/200943235/ partner/20094323 5
                  >
                  > Its a simple 2d mathematical function of a sine with increasing frequency.
                  >
                  > Best regards
                  >
                  >
                  >
                  > ____________ _________ _________ __
                  > From: Ken Warner <kwarner000@verizon. net>
                  > To: PanoToolsNG@ yahoogroups. .com
                  > Sent: Monday, April 6, 2009 2:04:13 AM
                  > Subject: [PanoToolsNG] I need a test image to show the difference between Lanczos and BiCubic iterpolation. ..
                  >
                  >
                  > I've written my own Lanczos and bicubic interpolators. When I apply
                  > them to the typical equirectangular pano, I really don't see much difference.
                  >
                  > I'm trying to talk myself into believing that the Lanczos interpolator is
                  > better than the bicubic interpolator but I can't. I just can't see the
                  > difference. In fact I can barely see the difference between bilinear
                  > and bicubic interpolation. Bilinear is only slightly softer.
                  >
                  > So what I need is a test image that I can use to show the difference
                  > to myself. Dr. Dersch used some images with concentric circles --
                  >
                  > http://home. no.net/dmaurer/ ~dersch/interpol ator/interpolato r.html
                  >
                  > Where does one get a test image like that?
                  >
                  > Or does anyone know another image that will show defects and clarity
                  > of interpolation?
                  >
                  >
                  > __._,_..___
                  > Messages in this topic (1) Reply (via web post) | Start a new topic
                  > Messages | Files | Photos | Links | Database | Polls | Members | Calendar






                  [Non-text portions of this message have been removed]
                • Ken Warner
                  Hi Fulvio, There are a bunch of different bicubic interpolation methods, kernals, basis functions, polynomials, or whatever you call them. The one I use -- and
                  Message 8 of 8 , Apr 7, 2009
                  • 0 Attachment
                    Hi Fulvio,

                    There are a bunch of different bicubic interpolation methods,
                    kernals, basis functions, polynomials, or whatever you call them.

                    The one I use -- and I don't know how to identify it -- has
                    one variable that I can change to make the basis functions produce
                    a sharper or softer interpolation.

                    I've read that either Lanczos or bicubic interpolation is better
                    for upscaling or downscaling but I forget which one is best applied
                    to which.

                    But at the resolution we work at on the web -- typically about
                    1k x 1k viewable and 8 bit per channel color -- I don't think it
                    really matters which method is used. Except for the ringing which
                    can be seen if sharpening is done too agressively.

                    For print, I'm sure that the differences would become apparent immediately.

                    There's other interpolation methods which I know nothing about except
                    that they do edge detection and interpolate the image in a way that
                    takes edge detection into account. The examples I've seen
                    look very good. I think I've seen the acronym NEDI associated
                    with those methods.

                    Ken

                    Fulvio Senore wrote:
                    > You can see some example in this page:
                    >
                    > http://www.fsoft.it/panorama/FSPViewer.htm
                    >
                    > there are bilinear and some Lanczos ones, no bicubic, sorry.
                    >
                    > Bicubic is a rather generic term, but some time ago I looked at the
                    > Lanczos and bicubic implementations in PTGui and I believe that bicubic
                    > normal is better.
                    > Lanczos has a slightly sharpening effect that can cause ringing (echoes
                    > of sharp edges, like in overcompressed jpegs).
                    > For this reason IMHO Lanczos is better for viewers where a sharp image
                    > is more pleasant, but bicubic is better for file creation, where the
                    > lanczos oversharpening might not be desired.
                    >
                    > About the interpolators test in the Dersch page, I think that it is not
                    > representative of real usage. Nobody applies so many interpolations to
                    > the same image.
                    > That test shows that a large windowed sinc kernel works very well, but
                    > this is not always true. Just try to interpolate a screenshot with sharp
                    > edges and you will see large ringing effects.
                    >
                    > Fulvio Senore
                    >
                    >
                    > Ken Warner ha scritto:
                    >
                    >>I've written my own Lanczos and bicubic interpolators. When I apply
                    >>them to the typical equirectangular pano, I really don't see much difference.
                    >>
                    >>I'm trying to talk myself into believing that the Lanczos interpolator is
                    >>better than the bicubic interpolator but I can't. I just can't see the
                    >>difference. In fact I can barely see the difference between bilinear
                    >>and bicubic interpolation. Bilinear is only slightly softer.
                    >>
                    >>So what I need is a test image that I can use to show the difference
                    >>to myself. Dr. Dersch used some images with concentric circles --
                    >>
                    >>http://home.no.net/dmaurer/~dersch/interpolator/interpolator.html
                    >>
                    >>Where does one get a test image like that?
                    >>
                    >>Or does anyone know another image that will show defects and clarity
                    >>of interpolation?
                    >>
                    >>
                    >>
                    >>------------------------------------
                    >>
                    >
                    >
                  Your message has been successfully submitted and would be delivered to recipients shortly.