forums.naturalpoint.com

I've been examining the data coming out of the Rigid Body Toolkit a bit more closely for a simple example with three markers (the marker-set typical rigid body) on a rigid body. I stuck the rigid body to the back of a rotating chair in plain sight of all three cameras - within 2 meters. With the markers in full view for the entire trial I rotate the chair +/- roughly 40 degrees (heading angle starts at 0 and rotates back and forth slowly). Even with the chair at rest with a heading of nearly 0, I get intermittent rigid body tracking in the exported csv file. I copied the first few frames of data from that file to the end of this post to illustrate the problem. Check out frames 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19 - all of which did not find a rigid body! Is this common or could I be doing something wrong/have a bad setting somewhere? When I plot the data (either quaternions or Heading/Attitude/Bank), I see the gaps but the data points that are there look to be relatively smooth and accurate. I tried re-calibrating with different threshold/exposure settings trying to ensure that the markers weren't "flickering" and I believe things are actually OK. When I review the same trial in the Rigid Body Toolkit, things look fine. I will email the calibration file, the mkr file, and the csv file for you to review.
Thanks in advance,
Brian

rigidbody Rigid Body 1 1 3 -0.08032852 0.01959178 0.01201609 0.04118107 -0.04259028 -0.00862135 0.03914744 0.02299848 -0.00339471
frame 0 1 1 0.10070401 0.27870545 0.1326689 0.00041868 0.00011336 -0.00016542 0.99999988 3 0.02038444 0.29833609 0.14472136 0.1398491 0.30172676 0.1292883 0.14189477 0.23612683 0.12400055
frame 1 1 1 0.10070609 0.27871376 0.13265775 0.00085669 -0.00047144 -0.00018993 0.99999958 3 0.02036837 0.29832697 0.14463326 0.1398593 0.30172843 0.12934212 0.14189477 0.23612683 0.12400055
frame 2 0 3 0.02036906 0.29834628 0.14462231 0.13982442 0.30170149 0.12933527 0.14188272 0.23611312 0.12391242
frame 3 0 3 0.02038524 0.29833639 0.14469701 0.13981661 0.30169225 0.12922691 0.14190784 0.23612194 0.12409955
frame 4 1 1 0.10069934 0.27871341 0.13265382 0.00097912 -0.00060441 -0.00020307 0.9999994 3 0.0203631 0.2983076 0.14461081 0.13984156 0.30173621 0.1293565 0.14187233 0.23611049 0.12399801
frame 5 0 3 0.02036837 0.29832697 0.14463326 0.1398192 0.30164957 0.12923956 0.1418605 0.23613775 0.12399648
frame 6 0 3 0.02037915 0.29829764 0.14468482 0.13983147 0.30171657 0.1292682 0.14188787 0.23613577 0.12394723
frame 7 0 3 0.02036837 0.29832697 0.14463326 0.13989034 0.30170903 0.12922931 0.14189477 0.23612683 0.12400055
frame 8 1 1 0.10070004 0.27869937 0.1326627 0.00097407 -0.00053133 -0.00023019 0.9999994 3 0.02036837 0.29832697 0.14463326 0.13985658 0.30171013 0.12935691 0.14189477 0.23612683 0.12400055
frame 9 0 3 0.02037816 0.29829895 0.14471623 0.13985698 0.30170947 0.12935993 0.14189477 0.23612683 0.12400055
frame 10 0 3 0.02036892 0.29836547 0.14463559 0.13987592 0.30172676 0.12936011 0.14187233 0.23611049 0.12399801
frame 11 1 1 0.10071199 0.27870056 0.13265179 0.00042189 0.00013544 -0.00017172 0.99999994 3 0.02038456 0.2983171 0.14470792 0.13985109 0.3017332 0.12927133 0.14189152 0.23609781 0.1239809
frame 12 0 3 0.02038509 0.29835558 0.14471027 0.13987726 0.301705 0.1292173 0.14189477 0.23612683 0.12400055
frame 13 0 3 0.02036147 0.2982896 0.14465302 0.13984156 0.30173621 0.1293565 0.14190356 0.23611133 0.12406912
frame 14 1 1 0.10069489 0.2786828 0.13267288 0.00051725 0.00006322 -0.00021567 0.99999994 3 0.02038444 0.29833609 0.14472136 0.13984838 0.30169141 0.12929927 0.14189477 0.23612683 0.12400055
frame 15 0 3 0.02032962 0.29833272 0.14466023 0.1398665 0.30170643 0.12927477 0.14190531 0.2361152 0.12408179
frame 16 0 3 0.02038509 0.29835558 0.14471027 0.13985658 0.30171013 0.12935691 0.14189477 0.23612683 0.12400055
frame 17 1 1 0.10071539 0.27873316 0.13267896 0.0010625 -0.00038537 -0.00018851 0.99999934 3 0.02036675 0.29830897 0.14467552 0.13987631 0.30172607 0.12936315 0.14187233 0.23611049 0.12399801
frame 18 0 3 0.02038456 0.2983171 0.14470792 0.13985318 0.30167675 0.12925446 0.14189477 0.23612683 0.12400055
frame 19 0 3 0.02037923 0.29830936 0.14468424 0.13983703 0.30167991 0.12930839 0.14190131 0.23608768 0.1240564
frame 20 1 1 0.10068884 0.27868921 0.13265923 0.00087857 -0.00048236 -0.00022315 0.99999952 3 0.02036837 0.29832697 0.14463326 0.13982253 0.30173832 0.12935106 0.14189477 0.23612683 0.12400055

I submitted the calibration file, the marker file, and the exported text file late on 8/8 to 8/9 but haven't heard anything back on this post. Anybody have suggestions regarding why the exported data file drops the rigid body periodically?
Brian

The data has been fowarded onto the development team, but they currently are at the Siggraph trade show.

I have the same problem as Brian and I think it is a problem of how the data are exported to the .csv file. I captured some data in .mrk format (the native format of Optitrak) and then I exported them to a .csv file. When I loaded the .mrk file and compared the results to those of the .csv file I found out that for a large number of frames where the .mrk file shows a Rigid Body, there is no such in the .csv file. I mean, the .mrk file shows that the Rigid Body tracking has been properly done, but when you export the data to .csv it fails.

By the way, it would be a good idea to allow the user to select a specific frame in the Optitrack Rigid Body Tool to check these kind of things. Now there is a bar that does not allow any accurate search. It would be also interesting to show the coordinates of each one of the markers for every frame. Now it just shows those of the initial position.

Thanks in advance for any help or update of the software,

Daniel

This particular issue has been resolved and will be fixed in the next release. If you need this fix before then, email support and they'll send you a nightly build of the installer for the Tracking Tools and the OptiTrack SDK.

The Rigid Body Toolkit is undergoing some large improvements that will be in a major release in the months to come.

Thank you Doug - about when should I look for that next release?
Brian