Skip to content


Subversion checkout URL

You can clone with
Download ZIP
Branch: master
Commits on Mar 31, 2015
  1. Compile fixes.

    Edward Rosten authored
Commits on Jan 16, 2014
  1. Fix for Mavericks from Clemens Arth.

    Edward Rosten authored
Commits on Aug 13, 2013
  1. Fix VS2012 test from C. Arth.

    Edward Rosten authored
  2. Add in warning for VS2012 compile errors

    Edward Rosten authored
    This is to do with the maximum tuple size.
Commits on Jan 7, 2013
  1. fixed a copy&paste bug in absorient. luckily it didn't change the res…

    Gerhard Reitmayr authored
    …ults, only potentially the accuracy.
Commits on Dec 6, 2012
  1. SVD.h needs to be caps in tag/absorient.h

    Paul McIlroy authored
Commits on Aug 1, 2011
  1. fix for different tr1 header locations in VS

    gerhard authored
Commits on Jul 28, 2011
  1. fix class/struct mismatch

    edrosten authored
Commits on Jun 30, 2011
  1. updated to TooN::SIM23 types

    gerhard authored
Commits on Jun 29, 2011
  1. made the N-dim absolute pose algorithm accessible, use new SIM{2,3} o…

    gerhard authored
    …bjects from TooN for 2D and 3D
Commits on Jun 27, 2011
  1. updated absolute orientation functions with a different method to cal…

    gerhard authored
    …culate the rotation and made them more robust against input with too little constraints as well
Commits on Jun 25, 2011
  1. clearer output

    gerhard authored
  2. added estimating a similarity transform (se3 + scale) between two 3D …

    gerhard authored
    …point sets after Horn
Commits on Nov 17, 2010
Commits on Jun 17, 2010
  1. added interface for three point pose to allow using 3 rays instead of…

    qpan authored
    … 3 perspective projected points
Commits on May 6, 2010
  1. small doc fix

    gerhard authored
  2. moved Kalman filter framework to the new age of TooN 2. Fixed all the…

    gerhard authored
    … TooN 1.0isms, different update implementation using Cholesky for better stability (keeps things symmetric) and no dependency on transformCovariance anymore (a small performance penality in predict ...)
  3. fixes by Ryan Pavlik

    gerhard authored
Commits on Jan 19, 2010
Commits on Jan 13, 2010
  1. Fixed to work with TooN2

    qpan authored
Commits on Nov 17, 2009
  1. Replace broken (?) build files?

    edrosten authored
Commits on Aug 25, 2009
  1. Put fourpointpose back in to the build.

    edrosten authored
  2. TooN-2 updates to four point pose.

    edrosten authored
Commits on Aug 15, 2009
  1. doc refinement

    gerhard authored
Commits on Aug 7, 2009
  1. added missing project file

    gerhard authored
Commits on Aug 5, 2009
  1. docu on win32 build files

    gerhard authored
  2. - update to build system to include an install project

    gerhard authored
    - added drand48 implementation for Win32
Commits on Jul 15, 2009
  1. Replace pair<A, B> with tuple<A, B> so that tie() works with older

    edrosten authored
    standard libraries.
Commits on Jul 12, 2009
  1. fixes for MS VC 2008 (VC9).

    gerhard authored
    Yes, it supports TR1, and but the headers are not in <tr1/...> ???
Commits on May 15, 2009
  1. 4x speedup to five_point. The new code will give slightly different r…

    edrosten authored
    to the old code, due to differences in numerics, tolerances and etc.
    The nullspace is now found using Gauss-Jordan, instead of SVD (for a speedup
    of about 30x for this particular part).
    New hand written polynomial root finder, for a speedup of 4x and improved
    reliability (the old one would occasionally expect -1 roots due to numerical
    instabilities and fail with an alloc_error).
    The new root-finder will not attempt to find roots of p(x) for |x|>100, since
    it is not possible to find these root to high enough accuracy.
    Roots are bracketed using Sturm sequences (as before) but using the resursive
    implementation which is more efficient.
    The root polishing uses a hybrid Newton-Raphson/bisection rootfinder which
    usually gets roots to higher accuracy than the old bisection solver and almost
    always converges in < 8 iterations, and often in 4. It will give up after 10
    A new tolerance feps is introduced, so that for a numerical root at x, if
    |p(x)| > feps, the root will not be recorded.
    The other tolerances (eps and zeps) follow the convention of NR in C.
Commits on May 11, 2009
  1. Fixed a very silly bug.

    edrosten authored
  2. Doc punt.

    edrosten authored
    Make essential_reprojection_errors_squared consistent wuith five_point
    in the argument ordering.
  3. Doc fix

    edrosten authored
  4. Reprojection error for the essential matrix.

    edrosten authored
Commits on May 8, 2009
  1. more TooN2 cleanup

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