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

smartblend

Expand Messages
  • jeffrey
    Has anyone heard from Michael Norel lately? http://smartblend.panotools.info/ is the home of smartblend. Can t we do better than that? ;) I hope Michael is
    Message 1 of 9 , Oct 30, 2006
    • 0 Attachment
      Has anyone heard from Michael Norel lately?

      http://smartblend.panotools.info/ is the home of smartblend.

      Can't we do better than that? ;)

      I hope Michael is still working on this valuable tool! I'm still not
      able to render huge files, but otherwise, it's a great weapon!!

      jeffrey
    • Erik Krause
      ... Yes, me. I sent him a bugreport 3 weeks ago, he replied he found the problem and asked me whether I could do a test with a 30,000x15,000 pixel pano. I
      Message 2 of 9 , Oct 30, 2006
      • 0 Attachment
        On Monday, October 30, 2006 at 20:28, jeffrey wrote:

        > Has anyone heard from Michael Norel lately?

        Yes, me. I sent him a bugreport 3 weeks ago, he replied he found the
        problem and asked me whether I could do a test with a 30,000x15,000
        pixel pano. I couldn't until now, but if anyone whishes to do the
        test I can provide the debug version of smartblend 1.2.3 he gave me.
        Please contact me off list...

        > http://smartblend.panotools.info/ is the home of smartblend.
        >
        > Can't we do better than that? ;)

        Yes, definitely: http://wiki.panotools.org/SmartBlend
        And of course we should convince Mike to leave panotools.info...

        best regards
        --
        Erik Krause
        Resources, not only for panorama creation:
        http://www.erik-krause.de/
      • JD Smith
        ... Can someone summarize where and why SmartBlend is preferable to Enblend (and where it s the other way around)? It would be nice if the features of
        Message 3 of 9 , Oct 30, 2006
        • 0 Attachment
          On Tue, 31 Oct 2006 00:04:37 +0100, Erik Krause wrote:

          > Yes, me. I sent him a bugreport 3 weeks ago, he replied he found the
          > problem and asked me whether I could do a test with a 30,000x15,000
          > pixel pano. I couldn't until now, but if anyone whishes to do the
          > test I can provide the debug version of smartblend 1.2.3 he gave me.
          > Please contact me off list...
          >
          >> [quoted text muted]
          >
          > Yes, definitely: http://wiki.panotools.org/SmartBlend
          > And of course we should convince Mike to leave panotools.info...

          Can someone summarize where and why SmartBlend is preferable to
          Enblend (and where it's the other way around)? It would be nice if
          the features of Smartblend which made it desirable migrated into
          enblend, which is open source and thus won't suffer bit rot.

          JD
        • panovrx
          JD asks ... Smartblend will stitch around action areas and warp images to fit. The only time I wouldnt use it is sometimes when I am stitching two sets of
          Message 4 of 9 , Oct 30, 2006
          • 0 Attachment
            JD asks
            > Can someone summarize where and why SmartBlend is preferable to
            > Enblend (and where it's the other way around)?

            Smartblend will stitch around action areas and warp images to fit. The
            only time I wouldnt use it is sometimes when I am stitching two sets of
            exposures extracted from the same set of RAW files -- where it will
            sometimes warp the two lots of stitching differently -- so the
            resultant panoramas no longer align. There is maybe some way of telling
            it to stitch both with the same warping/selections but I dont what it
            is. It is kinda slow with 16bit images.

            Peter M
            http://www.mediavr.com/blog
          • Tim Hatch
            ... I thought this too, but then I tried enblend with 16bit and it took a long time too. I think Smartblend may take a tad longer but is also a bit more
            Message 5 of 9 , Oct 30, 2006
            • 0 Attachment
              > It is kinda slow with 16bit images.

              I thought this too, but then I tried enblend with 16bit and it took
              "a long time" too. I think Smartblend may take a tad longer but is
              also a bit more upfront about it so you can groan at the beginning of
              the blend step instead of five minutes into it when it's done half
              done yet.

              Tim
            • Erik Krause
              ... Well, not exactly warp. It doesn t alter image geometry. It does a mere multiresolution spline blending like enblend does with the additional advantage
              Message 6 of 9 , Oct 31, 2006
              • 0 Attachment
                On Tuesday, October 31, 2006 at 1:59, panovrx wrote:

                > JD asks
                > > Can someone summarize where and why SmartBlend is preferable to
                > > Enblend (and where it's the other way around)?
                >
                > Smartblend will stitch around action areas and warp images to fit.

                Well, not exactly warp. It doesn't alter image geometry. It does a
                mere multiresolution spline blending like enblend does with the
                additional advantage that it sets the seam lines such that ghosting
                is avoided.

                And it seems as if the blend region is usually larger than that of
                enblend. Even larger than with the largest value allowed for the -l
                parameter in enblend. This results in smoother skies f.e.

                JD wrote:
                > It would be nice if the features of Smartblend which made it desirable
                > migrated into enblend, which is open source and thus won't suffer bit rot.

                Yes of course. Mike is unavailable sometimes and the smartblend
                development is far behind it's possibilities. It took him ages to
                implement larger images and 16 bit support where enblend has
                gigapixel support and 32bit float (HDR) already.

                best regards
                --
                Erik Krause
                Resources, not only for panorama creation:
                http://www.erik-krause.de/
              • JD Smith
                ... So, if the seam line algorithm in enblend is improved (I believe Andrew is working on this now) to avoid ghosts, SmartBlend would not have any strong
                Message 7 of 9 , Oct 31, 2006
                • 0 Attachment
                  On Tue, 31 Oct 2006 10:32:31 +0100, Erik Krause wrote:

                  > On Tuesday, October 31, 2006 at 1:59, panovrx wrote:
                  >
                  >> JD asks
                  >> > Can someone summarize where and why SmartBlend is preferable to
                  >> > Enblend (and where it's the other way around)?
                  >>
                  >> Smartblend will stitch around action areas and warp images to fit.
                  >
                  > Well, not exactly warp. It doesn't alter image geometry. It does a
                  > mere multiresolution spline blending like enblend does with the
                  > additional advantage that it sets the seam lines such that ghosting
                  > is avoided.

                  So, if the seam line algorithm in enblend is improved (I believe Andrew
                  is working on this now) to avoid ghosts, SmartBlend would not have any
                  strong advantage any longer?

                  JD
                • Fulvio Senore
                  ... Another advantage is speed. I have been using smartblend for a long time also because its speed. Yesterday I tried again enblend to make a comparison with
                  Message 8 of 9 , Oct 31, 2006
                  • 0 Attachment
                    JD Smith ha scritto:
                    > On Tue, 31 Oct 2006 10:32:31 +0100, Erik Krause wrote:
                    >
                    >
                    >> On Tuesday, October 31, 2006 at 1:59, panovrx wrote:
                    >>
                    >>
                    >>> JD asks
                    >>>
                    >>>> Can someone summarize where and why SmartBlend is preferable to
                    >>>> Enblend (and where it's the other way around)?
                    >>>>
                    >>> Smartblend will stitch around action areas and warp images to fit.
                    >>>
                    >> Well, not exactly warp. It doesn't alter image geometry. It does a
                    >> mere multiresolution spline blending like enblend does with the
                    >> additional advantage that it sets the seam lines such that ghosting
                    >> is avoided.
                    >>
                    >
                    > So, if the seam line algorithm in enblend is improved (I believe Andrew
                    > is working on this now) to avoid ghosts, SmartBlend would not have any
                    > strong advantage any longer?
                    >
                    > JD
                    >
                    Another advantage is speed. I have been using smartblend for a long time
                    also because its speed.

                    Yesterday I tried again enblend to make a comparison with a pano shoot
                    handheld among the crowd holding a camera with a 28mm eq lens over my head.
                    Enblend has been so slow that I found it almost ridicolus. I can get
                    better result in a dramatically shorter time with smartblend.

                    Smartbend ran in 1 minute and 37 seconds, enblend ran in 14 minutes and
                    20 seconds. No, I did not make any typing mistake.

                    Fulvio Senore
                  • Erik Krause
                    ... Except that I can only think of the smoother blending in uniform areas (f.e. sky). And there might be a speed advantage for not so large 8 bit images...
                    Message 9 of 9 , Oct 31, 2006
                    • 0 Attachment
                      On Tuesday, October 31, 2006 at 13:40, JD Smith wrote:

                      > > Well, not exactly warp. It doesn't alter image geometry. It does a
                      > > mere multiresolution spline blending like enblend does with the
                      > > additional advantage that it sets the seam lines such that ghosting
                      > > is avoided.
                      >
                      > So, if the seam line algorithm in enblend is improved (I believe Andrew
                      > is working on this now) to avoid ghosts, SmartBlend would not have any
                      > strong advantage any longer?

                      Except that I can only think of the smoother blending in uniform
                      areas (f.e. sky). And there might be a speed advantage for not so
                      large 8 bit images...

                      best regards
                      --
                      Erik Krause
                      Resources, not only for panorama creation:
                      http://www.erik-krause.de/
                    Your message has been successfully submitted and would be delivered to recipients shortly.