Luca
cd6c5ca0bd
using parameters in smart projection factors constructors.. breaking the API, but now is way more elegant
2015-06-19 18:09:39 -04:00
Luca
f8205bfe02
Merge branch 'develop' into feature/SmartFactors3
...
Conflicts:
gtsam/geometry/CalibratedCamera.h
gtsam/geometry/tests/testCalibratedCamera.cpp
gtsam/geometry/triangulation.h
gtsam/nonlinear/ExpressionFactor.h
gtsam/slam/SmartProjectionFactor.h
gtsam_unstable/slam/SmartStereoProjectionFactor.h
tests/testExpressionFactor.cpp
2015-06-19 16:55:30 -04:00
Luca
2d9fddbcaa
made cameras virtual and simplified pose factor (with Frank)
2015-06-19 15:33:18 -04:00
Luca
dcce21639f
included new constructors (still commented out)
2015-06-19 14:59:33 -04:00
Luca
f1e5c61762
adding parameter structure
2015-06-19 13:10:42 -04:00
Luca
917e7c177d
fixed examples
2015-06-19 13:10:34 -04:00
Luca
78c8160dc5
all tests pass and it compiles (yuppii!), but if I make check I get errors with isManifold and something that seems unrelated to smart factors. going to merge with develop
2015-06-19 12:06:45 -04:00
Luca
fb7bc12c84
most unit tests fixed - 2 to go, now sensor_T_body is in the base class, probably not the best choice
2015-06-19 11:42:51 -04:00
Luca
100016e3af
put sensor_T_body in SmartProjectionFactor
2015-06-19 10:39:59 -04:00
Luca
756d1d29b7
fixing key unit tests - still failures in the optimization
2015-06-18 17:23:39 -04:00
Luca
66083f5e18
included calibration in constructor SmartProjectionPoseFactor
2015-06-18 16:33:17 -04:00
Luca
391386a654
revived SmartProjectionPoseFactor. Compiles now, going to fix unit tests now
2015-06-18 16:00:54 -04:00
Frank Dellaert
9d40113dda
Snavely does not work
2015-06-18 09:56:04 -07:00
Frank Dellaert
c01618a7ff
Use Ceres defaults
2015-06-18 09:51:30 -07:00
Frank Dellaert
879e66a63a
TWo default param sets
2015-06-18 09:51:22 -07:00
Frank Dellaert
9f209e441f
Merge branch 'fix_constructor' into feature/CeresDefaults
2015-06-17 19:54:34 -07:00
Frank Dellaert
62bfce7358
Fixed state constructor to set reuseDiagonal properly
2015-06-17 19:54:21 -07:00
Frank Dellaert
62db9370ca
Better separation of cers/gtam paths
2015-06-17 19:43:26 -07:00
Frank Dellaert
ba932cafae
Spelling
2015-06-17 19:43:04 -07:00
Chris Beall
cfa656c2b9
Merged in feature/prohibitTimingTBB (pull request #158 )
...
Prohibit Timing build mode with TBB.
2015-06-17 17:00:36 -04:00
Frank Dellaert
e7d10b8080
Merged in fix/zeroRowInQR (pull request #155 )
...
Change in QR factorization
2015-06-17 13:59:28 -07:00
cbeall3
0f02b7d473
Prohibit Timing build mode with TBB. See issue #173
2015-06-17 16:23:27 -04:00
dellaert
1269785c05
Fixed naming convention
2015-06-17 09:23:24 -07:00
dellaert
d71e66ea48
Moved reuse_diagnal_ to reuseDiagonal in state
2015-06-17 09:20:46 -07:00
Chris Beall
19bc1cd686
Merged in fix/sparse_jacobian_output (pull request #121 )
...
fixed sparseJacobian() to use a std::map such that it works with symbol keys
2015-06-17 12:17:59 -04:00
dellaert
6efc708c5e
For exact comparison with Ceres, use exact same AutoDiff model
2015-06-17 09:06:21 -07:00
dellaert
178c1aec88
Better comment
2015-06-17 09:05:02 -07:00
Frank Dellaert
a9339baef0
Merged develop into fix/sparse_jacobian_output
2015-06-17 08:37:55 -07:00
Frank Dellaert
32044eaac8
Added a named constructor to mimick Ceres defaults
2015-06-16 22:24:39 -07:00
Frank Dellaert
b91ed6dc6e
Got rid of warning
2015-06-16 21:17:22 -07:00
Frank Dellaert
ca2aa0cfd4
Refactoring to get grouped factors to compile again
2015-06-16 20:33:33 -07:00
dellaert
82f6ca64a5
Merge remote-tracking branch 'origin/develop' into fix/zeroRowInQR
2015-06-15 01:09:19 -07:00
Frank Dellaert
b68f763fe7
Merged in feature/LinearSolverCleanup (pull request #148 )
...
This has
- Separate Scatter class
- change from updateATA to updateHessian as a virtual method
- Fixed-size BinaryJacobianFactor that overloads updateHessian
- SFM Timing Script that takes BAL files
2015-06-15 01:08:02 -07:00
dellaert
33e412f2ee
Code review
2015-06-15 01:05:48 -07:00
dellaert
f523db2d98
IMPORTANT CHANGE/BUGFIX: QR elimination now (I think properly) leaves a row of zeros if the RHS after QR contains a non-zero. A corresponding change in the error calculation makes that Jacobian and Hessian factors now agree on error. I think this was a bug, because it affected the error, but (I think) it only pertained to "empty" JacobianFactors which have no bearing on optimization/elimination.
2015-06-15 00:55:53 -07:00
dellaert
83171b6096
New example
2015-06-14 23:07:28 -07:00
Frank Dellaert
51433c1ec2
Merged in feature/cleanupBase (pull request #154 )
...
Some cleanup in base
2015-06-14 20:42:13 -07:00
dellaert
75e072396c
Refactored and renamed some internals
2015-06-14 16:05:39 -07:00
dellaert
15966a65f2
Small reformat
2015-06-14 16:05:18 -07:00
dellaert
a18875b598
Changed headers to GTSAM-style (near to far)
2015-06-14 16:05:09 -07:00
dellaert
5771664622
Starting to diagnose issue with lower-left entry of Hessian
2015-06-14 16:01:54 -07:00
dellaert
7698c52ce9
Created BinaryJacobianFactor template
2015-06-14 15:50:15 -07:00
dellaert
06902209b0
Fixed bug in hessian scalar computation
2015-06-14 15:19:55 -07:00
dellaert
30104a114e
More tests with failing example
2015-06-14 15:03:44 -07:00
dellaert
a94c2e7323
Renamed to BinaryJacobianFactor
2015-06-14 15:02:48 -07:00
dellaert
9fcd498d6a
BORG formatting
2015-06-14 13:37:51 -07:00
dellaert
8c22684bbb
Went back to base 1, and used constructors for blocks (cleaner)
2015-06-14 13:02:44 -07:00
dellaert
df226fc436
No longer store my own matrices but get same performance using block
2015-06-14 12:54:18 -07:00
dellaert
e045a5e1f7
Added more powerful tests on updateHessian
2015-06-14 12:53:20 -07:00
dellaert
850501ed52
BORG Formatting
2015-06-14 11:16:54 -07:00