Skip to content
This repository

HTTPS clone URL

Subversion checkout URL

You can clone with HTTPS or Subversion.

Download ZIP

Mar 07, 2012

  1. Chris

    more bug_fixes in fit_planes, angle dist is good now, only legitimate…

    …ly bad planes are coming back as unpossible
  2. Chris

    still debuging, angles seem to be mostly sane now, still quite a few …

    …planes ~ 20 were nans
  3. Chris

    added zero-clamp to angle calc in fit_planes, this rounds small value…

    …s back to zero to stop weird problems with averages in the modular arith
  4. Chris

    fixed vert/horiz classification bug, using normals not centers to do …

    …classification so should be somewhat immune to scale

Feb 27, 2012

  1. Chris

    output the empirical centroid if theres no planes possible, angle = 0

    authored
  2. Chris

    fitCubes.py updated to use either fitplanes or icp_cubes

    authored
  3. Chris

    moved estimated cube resolutions into constants.h, cleaned up cmakefi…

    …les a bit
    authored
  4. Chris

    moved cubeShortSide and cubeLongSide into constants.h

    authored
  5. Chris

    fit planes finally works for just horizontal planes too, all good in …

    …the hood
    authored

Feb 26, 2012

  1. Chris

    testcubes are now built with the correct number of points, theyre now…

    … actually useful for testin again
    authored
  2. Chris

    fixed arg counting error, removed some unused variables

    authored

Feb 20, 2012

  1. Chris

    fit planes debugged somewhat, works quite well with mixed cubes, but …

    …some normals are coming out weird in the no-horiztonals or no-verticals case, look into this
    authored
  2. Chris

    tidying up source info

    authored
  3. Chris

    removed useless convex-hull example

    authored
  4. Chris

    test cubes will make full and partial cubes

    authored
  5. Chris

    make_test_cubes now outputs test with some translation adn rotation, …

    …they look nice
    authored
  6. Chris

    icp_cubes should produce all output, added a note to switch to using …

    …pcl min-max finding code
    authored
  7. Chris

    fixed pts in cluster_extract

    authored
  8. Chris

    make_test_cubes is to check fit_planes in various configs

    authored
  9. Chris

    fit_planes computes centroid and angles, needs a little testing but i…

    …s pretty much squarely ok now
    authored
  10. Chris

    fit_planes actually does compute the centroid

    authored

Feb 15, 2012

  1. Chris

    fit_planes is having some weird segfault issues on laptop

    authored

Feb 12, 2012

  1. Chris

    saving transformations, pcd files and not cbin files anymore

    authored
  2. Chris

    long term fiddling full_cloud_register is very slow, probably doesnt …

    …work
    authored
  3. Chris

    fixed that horrible boost library bug, ened to include the boost stuf…

    …f BEFORE the pcl stuff or else something gets borked and the build wont work
    authored

Feb 11, 2012

  1. Chris

    updated readme, figured out how to build without using the broken boo…

    …st string build
    authored
  2. Chris

    renamed ransac to icp, since were using icp to do the alignment, seem…

    …s to work pretty wonderfully
    authored
  3. Chris

    moved source into its own folder

    authored
Something went wrong with that request. Please try again.