summaryrefslogtreecommitdiffhomepage
path: root/opentrack/tracker.cpp
AgeCommit message (Collapse)Author
2014-12-18implement camera offsetStanislaw Halik
2014-12-15move rmat->euler and euler->rmat conversions to headerStanislaw Halik
2014-11-29implement zero shortcutStanislaw Halik
2014-11-28don't reset coords on pauseStanislaw Halik
Issue: #99
2014-11-26tracker: map before filteringStanislaw Halik
2014-11-11sane gimbal lock boundsStanislaw Halik
Tested-by: @KyokushinPL Issue: #63
2014-11-11add again gimbal lock workaround codeStanislaw Halik
@KyokushinPL reports +-90 pitch boundary broken otherwise
2014-11-11centering changesStanislaw Halik
2014-11-10simplify centeringStanislaw Halik
2014-11-09guard against gimbal lockStanislaw Halik
Issue: #63
2014-11-09doesn't hurt to use the commented-out formulaStanislaw Halik
2014-11-09Revert "fix typos"Stanislaw Halik
This reverts commit e74f029bf790342b1d3e0f11e5edf98c16946b08.
2014-11-09add commentStanislaw Halik
2014-11-09fix typosStanislaw Halik
2014-11-07fix r/t comp after mat changeStanislaw Halik
2014-11-07no need use temp array anymoreStanislaw Halik
2014-11-07nix unused constantStanislaw Halik
2014-11-07impl centering not confused on 90deg boundariesStanislaw Halik
atan2 returned wrong quadrants, as per <http://en.wikipedia.org/wiki/Atan2#Definition_and_computation> breakage on sign change. Issue: #63
2014-11-05Revert "center after mapping, not before"Stanislaw Halik
This reverts commit 2b0e342a04bad49c6bf59d388ed874145117f815. That was stupid.
2014-11-05center after mapping, not beforeStanislaw Halik
This makes octopus less spastic. Tested with joystick input. Issue: #63
2014-11-02use quaternions for centeringStanislaw Halik
@KyokushinPL says was the only version that worked. Let's give it a try. Goddamn issue: #63
2014-11-01fix typoStanislaw Halik
2014-11-01nix warnStanislaw Halik
2014-11-01revert centering changesStanislaw Halik
2014-11-01use simple-mat now opencv nixed from coreStanislaw Halik
2014-10-29centering formula got wrong againStanislaw Halik
2014-10-29r/t compensation broken, fixStanislaw Halik
2014-10-29angle order for yaw, pitch, rollStanislaw Halik
Issue: #63
2014-10-28poke centering with a stickStanislaw Halik
Issue: #63
2014-10-28linear algebra or bust (interconnect between axis)Stanislaw Halik
Issue: #63 Reported-by: many
2014-10-27guard against negative sleep amountStanislaw Halik
2014-10-27fix tracker sleep_ms to correctly compute overheadStanislaw Halik
2014-10-27Send origin pose to protocol on stoppingStanislaw Halik
Requested-by: @Antmax
2014-10-26rework accela, less nonsensical user-facing optionsStanislaw Halik
2014-10-26r/t comp wrong orderStanislaw Halik
2014-10-26inverting early caused wrong asymmetric axis selectionStanislaw Halik
2014-10-26swap axes only after mapping them, not beforeStanislaw Halik
@KyokushinPL mumbled something about Google Cardboard being unusable. If breaks existing use-cases, need reverting though.
2014-10-25swap axis to fix r/t compensationStanislaw Halik
Issue: #63
2014-10-25Change rot order tait-bryan -> matStanislaw Halik
Issue: 63
2014-10-25Basis change got lost, replace itStanislaw Halik
Issue: #63
2014-10-25fix sign, r/t compensationStanislaw Halik
Issue: #63
2014-10-25basis mapping seem to work nowStanislaw Halik
Only matrix -> euler conversion broken Issue: #63
2014-10-24Rotation basis algebra for centeringStanislaw Halik
Reported-by: @doveman months ago, many times Issue: #63 @dbaarda please confirm that- or -whether- it makes any sense. Issue: #86 It could be either worse or better than before. Please specify. Sadly, no time to plug the videos as camera input to PT as of yet. Signed-off-by: Stanislaw Halik <sthalik@misaki.pl>
2014-10-22center to filtered pose, not rawStanislaw Halik
2014-10-21fix toggle output zeros regressionStanislaw Halik
2014-10-19decruft moreStanislaw Halik
generated by cgit v1.2.3 (git 2.39.1) at 2024-11-17 21:51:34 +0000