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

Re: [PanoToolsNG] Enfuse workflow thought

Expand Messages
  • Bruno Postle
    ... Yes, use align_image_stack in the hugin distribution. This does everything you want including writing out the aligned images, it even outputs a .pto
    Message 1 of 5 , Dec 31, 2007
    • 0 Attachment
      On Mon 31-Dec-2007 at 10:11 -0800, Roger Howard wrote:
      >
      >I realize this is probably out of scope for the core enfuse tool but I'm
      >sure it's also a good idea. If I were to try to piece together a cross
      >platform, unattended auto-registration tool (such as can be found in
      >Photoshop's "Auto Align Layers"), where to start?
      >
      >- Autosift to generate control points ->
      >- PanoTools to optimize ->
      >- PTstitcher to render out results to separate images ->
      >- Enfuse to do its thing
      >
      >Anything more direct? Something obvious I'm missing? Is there an open
      >source image align tool that I can use to skip the first three steps

      Yes, use align_image_stack in the hugin distribution. This does
      everything you want including writing out the aligned images, it
      even outputs a .pto script so you can open the originals as a hugin
      project.

      Actually align_image_stack needs a bit of work to allow it to be a
      general purpose tool for this sort of thing, for instance it assumes
      rectilinear input images, but could easily be given a fisheye
      option.

      Here are the command-line options:

      Usage: align_image_stack [options] input files
      Valid options are:
      Modes of operation:
      -p file Output .pto file (useful for debugging, or further refinement)
      -a prefix align images, output as prefix_xxxx.tif
      -o output merge images to HDR, generate output.hdr),
      Modifiers
      -v Verbose, print progress messages
      -t num Remove all control points with an error higher than num pixels (default: 3)
      -f HFOV approximate horizontal field of view of input images, use if EXIF info not complete
      -m Optimize field of view for all images, execpt for first.
      Useful for aligning focus stacks with slightly different magnification.
      -c num number of control points to create between adjectant images (default: 200)
      -h Display help (this text)

      --
      Bruno
    • Roger Howard
      ... Bruno, I don t see this anywhere in the latest SVN build of Hugin on OSX - sorry, I didn t specify my platform in my previous email. Any idea if this ships
      Message 2 of 5 , Dec 31, 2007
      • 0 Attachment
        > Yes, use align_image_stack in the hugin distribution. This does
        > everything you want including writing out the aligned images, it
        > even outputs a .pto script so you can open the originals as a hugin
        > project.
        >
        > Actually align_image_stack needs a bit of work to allow it to be a
        > general purpose tool for this sort of thing, for instance it assumes
        > rectilinear input images, but could easily be given a fisheye
        > option.

        Bruno,

        I don't see this anywhere in the latest SVN build of Hugin on OSX -
        sorry, I didn't specify my platform in my previous email. Any idea if
        this ships as a self-contained binary in Hugin on OSX? Is this dependent
        on Mono/.Net? Sounds like just what I need.
      • Bruno Postle
        ... align_image_stack uses the internal hugin feature detection and control-point fine tuning, so it doesn t depend on any external tools like autopano-sift.
        Message 3 of 5 , Dec 31, 2007
        • 0 Attachment
          On Mon 31-Dec-2007 at 11:05 -0800, Roger Howard wrote:

          > > Yes, use align_image_stack in the hugin distribution. This does
          > > everything you want including writing out the aligned images, it
          > > even outputs a .pto script so you can open the originals as a hugin
          > > project.

          > I don't see this anywhere in the latest SVN build of Hugin on OSX
          > - sorry, I didn't specify my platform in my previous email. Any
          > idea if this ships as a self-contained binary in Hugin on OSX? Is
          > this dependent on Mono/.Net? Sounds like just what I need.

          align_image_stack uses the internal hugin feature detection and
          control-point fine tuning, so it doesn't depend on any external
          tools like autopano-sift. There should be no more problem compiling
          it than nona or fulla (also found in ./src/tools).

          If it doesn't compile with the OS X build system then this is an
          oversight, as it builds ok with cmake.

          --
          Bruno
        • Boris Abramovich
          You could also try hdrprep. http://www.luxal.eu/resources/hdr/hdrprep/tutorial.shtml and http://digilander.libero.it/bepperota/hdrprep/hdrprep_edited.html
          Message 4 of 5 , Dec 31, 2007
          • 0 Attachment
            You could also try hdrprep.
            http://www.luxal.eu/resources/hdr/hdrprep/tutorial.shtml
            and
            http://digilander.libero.it/bepperota/hdrprep/hdrprep_edited.html

            Boris

            Bruno Postle wrote:
            >
            > On Mon 31-Dec-2007 at 10:11 -0800, Roger Howard wrote:
            > >
            > >I realize this is probably out of scope for the core enfuse tool but I'm
            > >sure it's also a good idea. If I were to try to piece together a cross
            > >platform, unattended auto-registration tool (such as can be found in
            > >Photoshop's "Auto Align Layers"), where to start?
            > >
            > >- Autosift to generate control points ->
            > >- PanoTools to optimize ->
            > >- PTstitcher to render out results to separate images ->
            > >- Enfuse to do its thing
            > >
            > >Anything more direct? Something obvious I'm missing? Is there an open
            > >source image align tool that I can use to skip the first three steps
            >
            > Yes, use align_image_stack in the hugin distribution. This does
            > everything you want including writing out the aligned images, it
            > even outputs a .pto script so you can open the originals as a hugin
            > project.
            >
            > Actually align_image_stack needs a bit of work to allow it to be a
            > general purpose tool for this sort of thing, for instance it assumes
            > rectilinear input images, but could easily be given a fisheye
            > option.
            >
            > Here are the command-line options:
            >
            > Usage: align_image_stack [options] input files
            > Valid options are:
            > Modes of operation:
            > -p file Output .pto file (useful for debugging, or further refinement)
            > -a prefix align images, output as prefix_xxxx.tif
            > -o output merge images to HDR, generate output.hdr),
            > Modifiers
            > -v Verbose, print progress messages
            > -t num Remove all control points with an error higher than num pixels
            > (default: 3)
            > -f HFOV approximate horizontal field of view of input images, use if
            > EXIF info not complete
            > -m Optimize field of view for all images, execpt for first.
            > Useful for aligning focus stacks with slightly different magnification.
            > -c num number of control points to create between adjectant images
            > (default: 200)
            > -h Display help (this text)
            >
            > --
            > Bruno
            >
            >
          Your message has been successfully submitted and would be delivered to recipients shortly.