Autooptimiser

From PanoTools.org Wiki
Revision as of 00:01, 29 September 2010 by Jwatters (talk | contribs) (mention vig_optimize)
Jump to navigation Jump to search

autooptimiser is a smart optimiser replacement for PTOptimizer and vig_optimize, it is part of the hugin suite of tools for panoramic imaging.

It is not a 100% drop-in replacement for PTOptimizer as it has a slightly different interface and some extra features:

  • PTOptimizer will write the output of the optimisation into the input script file as 'o' lines, deleting any that already exist. This behaviour is a little strange, so autooptimiser writes the result to STDOUT or a new file, as is more usual for a command-line tool.
  • Not clobbering the original file means that there is no need to create 'o' lines, the output file is in the same format as the input with optimised values applied to the 'i' lines.
  • PTStitcher requires 'o' lines, so the output of autooptimiser isn't suitable for stitching with PTStitcher without further processing - Stitch these files with nona instead.

Additional features:

  • autooptimiser can perform pairwise optimisation instead of trying to optimise all rotation parameters at the same time, this means that there is no need to pre-position images before optimisation.
  • levelling the horizon in a set of images can be performed, this is the same functionality as the Straighten button in the hugin Preview window.
  • auto-align mode will pick an optimisation scheme suited to the distribution of control points in the project so there is no need to specify parameters to optimise.
  • autooptimiser can adjust the panorama field of view, dimensions and projection to suit the input images.
  • the field of view of input images can be specified, this is useful as the project files created by autopano and autopano-sift don't specify the field of view of the input photos.
  • autooptimiser can also perform photometric optimisation like the vig_optimize tool. Depending on the input images it selects automaticly these photometric parameters (exposure, white balance, vignetting, camera response curve) which should be optimised.