Commit Graph

5332 Commits (bfd334dd0ec0db000d193965ffa1afcc0f58648f)

Author SHA1 Message Date
Richard Roberts bfd334dd0e Added .pyc files to ignore 2014-01-06 21:19:53 -05:00
Alex Cunningham 96a4d80da4 Fixed some warnings 2014-01-04 20:35:19 -05:00
Alex Cunningham e05a2adca8 Merge branch 'fix_serialization' into develop 2014-01-04 20:14:10 -05:00
Alex Cunningham b08e889222 removed assertion that apparently fails in a bunch of places 2014-01-04 20:13:00 -05:00
Alex Cunningham dfb3e60171 Fixed issue with serialization by actively changing inf/nan values to zero in constrained noisemodel. Updated WhitenInPlace() and distance(). Apparently this caused a bunch of other unit tests to fail, though. 2014-01-04 20:01:05 -05:00
Alex Cunningham 54c1ec0dcb comments only - found the source of the problem 2014-01-04 19:04:34 -05:00
Alex Cunningham 4aea198c84 Marked failure cases for serialization in the linear test - get "input stream error" with the constrained noise model on text and XML serialization modes, but not binary 2014-01-04 16:29:25 -05:00
Alex Cunningham 56ff0b514c Changed ignore to include all build folders starting with /build* 2014-01-04 15:33:59 -05:00
Richard Roberts 90eca73010 Renamed LICENSE to LICENSE.BSD 2014-01-04 13:59:43 -05:00
Richard Roberts 1e06e7f4c8 Updates to README.md 2014-01-04 12:31:12 -05:00
Richard Roberts 7662059185 Working on markup README.md 2014-01-04 03:15:25 -05:00
Richard Roberts 599728022c Testing markdown in README 2014-01-04 02:54:22 -05:00
Frank Dellaert a3003a0736 Merge branch 'origin/release/2.4.0'
Mainly improvements in EssentialMatrix stuff, but also has the new precisions in the diagonal noise model.
2014-01-04 00:29:32 -05:00
Frank Dellaert d8b68c6fab Merged in feature/2.4.0/extrinsicE (pull request #2)
A third factor for essential matrices, now with an extrinsic calibration (rotation only)
2014-01-03 23:18:08 -05:00
Frank Dellaert fe3177c257 Extra constructor with calibration, unit tested 2014-01-03 23:12:04 -05:00
Frank Dellaert 4f81d110f1 Moved new class EssentialMatrixFactor3 into header 2014-01-03 22:22:48 -05:00
Frank Dellaert 861bd148e9 evaluateError Jacobians now implemented and unit tested 2014-01-03 22:17:05 -05:00
Frank Dellaert 5ae65d3f3a Derivatives of rotate work (at least, part that matters: not yet with respect to rotation) 2014-01-03 19:54:43 -05:00
Frank Dellaert b839387028 Evaluation works 2014-01-03 18:10:06 -05:00
Frank Dellaert 5002f3da8a Moved to header as friend function 2014-01-03 17:49:45 -05:00
Frank Dellaert d6864b0478 Rotate, without derivatives 2014-01-03 17:46:36 -05:00
Frank Dellaert 4b538d9648 Conjugation 2014-01-03 17:45:45 -05:00
Frank Dellaert 3afc4eb651 Merged in feature/2.4.0/precisions (pull request #1)
Better noise model for updateAtA using precisions
2014-01-03 17:23:05 -05:00
Frank Dellaert 18dfd70295 Removed excess baggage, and added templated constructor with calibration parameter 2014-01-03 16:08:38 -05:00
Frank Dellaert f54861f851 Made it more clear that Rotation constructor takes columns. 2014-01-03 15:56:45 -05:00
Frank Dellaert 29085b94a2 Check invsigmas and precisions indeed infinity in case of constrained models 2014-01-02 18:38:47 -05:00
Frank Dellaert 6f64b20a5b The precisions are used many many times in HessianFactor::updatATA, so I added them to the NoiseModel. The invsigmas are no longer optional and hence do not cause extra allocation at every invocation. Same with precision: returns const & 2014-01-02 18:05:44 -05:00
Frank Dellaert c394f9a601 Disabled non-working smartfactors test 2014-01-02 17:32:31 -05:00
Richard Roberts 7761796036 Path fix for projects referencing cmake libs without using a submodule 2013-12-29 15:13:13 -05:00
Richard Roberts 2805ecfe1c Allowing empty noise model in NonlinearFactor 2013-12-28 12:26:42 -05:00
Frank Dellaert 01dbca8ce1 Use Rot3 and Sphere2 random constructors to generate random E 2013-12-26 14:00:07 -05:00
Frank Dellaert a47f0a30da Use Sphere2::Random to generate a random Rot3 2013-12-26 13:59:44 -05:00
Frank Dellaert dc8236805f Generating random directions. Somewhat clumsy use of forward declaration, issue #19 applies as well. 2013-12-26 13:59:15 -05:00
Frank Dellaert 3b71fe47bc Merge branch 'origin/release/2.4.0' 2013-12-25 07:49:43 -05:00
Frank Dellaert e038fa4e6a Do calibrate of pA in constructor (needs to happen only once) 2013-12-24 17:26:13 -05:00
Frank Dellaert 606b9dce5c Fixed bug: first point (pA) had to be calibrated and it was not. 2013-12-24 17:14:28 -05:00
Frank Dellaert 707c745aad Created second example 2013-12-24 16:47:30 -05:00
Richard Roberts 6654fb2701 Fixed compiler error on MSVC 10.0 2013-12-24 14:49:01 -05:00
Frank Dellaert 886119bcf0 Merge branch 'origin/release/2.4.0' 2013-12-24 01:35:46 -05:00
Frank Dellaert 5f534aaa95 Merge branch 'origin/refineE' 2013-12-24 01:29:46 -05:00
Frank Dellaert e6a64ec5b2 Cleaned up derivatives somewhat more 2013-12-24 01:25:52 -05:00
Frank Dellaert adac9d209f Fixed unit test (baseline of test-case was not 1, corrected for it) 2013-12-24 01:25:13 -05:00
Frank Dellaert 708665ca04 Moved new factor to header; Still fails! 2013-12-24 00:52:24 -05:00
Frank Dellaert bba8368218 Added clone, minimization test; converges, but not to grund truth (which has high error) 2013-12-24 00:48:33 -05:00
Frank Dellaert d6edd1db07 Sanitized/collapsed derivatives 2013-12-24 00:36:54 -05:00
Frank Dellaert 44d03e638e Tedious derivatives, but right from the start! 2013-12-24 00:19:08 -05:00
Frank Dellaert 92c1398cd2 Checked evaluation and calculation 2013-12-23 22:29:09 -05:00
Frank Dellaert ce7cf524ea Initial unit test 2013-12-23 19:36:24 -05:00
Frank Dellaert 3b4fd765ec Merged wrap improvements back from develop branch 2013-12-23 14:41:49 -05:00
Frank Dellaert ec067b4954 gitinspector stats 2013-12-23 13:16:32 -05:00