initialize('web'); $snip = $modx->runSnippet("getSiteNavigation", array('id'=>5, 'phLevels'=>'sitenav.level0,sitenav.level1', 'showPageNav'=>'n')); $chunkOutput = $modx->getChunk("site-header", array('sitenav'=>$snip)); $bodytag = str_replace("[[+showSubmenus:notempty=`", "", $chunkOutput); $bodytag = str_replace("`]]", "", $bodytag); echo $bodytag; echo "\n"; ?>

Documentation

Evaluating pcl/registration

This is a collection of ideas on how to build an evaluation framework of pcl/registration.

Data generation

  • synthetic data
  • real word data (how to get ground truth?) - Kinect - PR2 laser scanner - SICK laser data - small range 3D scanner - mid range 3D scanner (Faro) - high end 3D scanner (Riegl, Velodyne)
  • Point Types - 2D(?) - 3D - RGB
  • dynamics - static scans - scanning while driving (e.g. robots)
  • size - room - building - outdoor (street)

Architecture

  • some lib for polygonal data
  • modeling different sensors
  • modeling noise
  • add a trajectory file
  • output a pile of .pcd files
  • integrate command line tools from PCL grandfather

Evaluating different algorithms

ICP

  • how does the algorithm cope with outliers
  • how are the point pairs evaluated:
    • does it use normal or RGB information
    • does it weight the pairs differently
    • which kind of point pairs are used:
      • one-to-one
      • one-to-many
      • many-to-many
getChunk("site-footer"); echo $chunkOutput; ?>