Good afternoon,
I would like to try to use OpenSim to capture and process gait data during load bearing. Specifically, we are studying head load carriage and infant carrying, and we hope to collect data outdoors. We started running some trials in OpenSim just to try it out and make sure everything works. We have had two consistent problems, even just with unloaded walking indoors. The first is that when we click the Analysis button for overground walking, it almost invariably fails. Out of 25+ trials, the Analysis has only been successful once. Do you know what might cause Analysis to fail? Second, we are having problems with calibration that we haven't been able to solve with the FAQ and best practices. Calibration is usually successful, but then in many videos it appears that the subject is levitating (i.e. the skeleton avatar is up to a foot or so off the ground surface), and this does seem to affect the kinematics results. Any light you can shed on these issues would be greatly appreciated!
Thank you, Laura Gruss
Analysis and calibration problems
-
Laura Gruss
- Posts: 3
- Joined: Sat Jul 06, 2024 5:40 am
Analysis and calibration problems
Last edited by Laura Gruss on Tue Jul 30, 2024 4:32 am, edited 1 time in total.
-
Laura Gruss
- Posts: 3
- Joined: Sat Jul 06, 2024 5:40 am
Re: Analysis and calibration problems
The bug fix solved the Analysis problem! Thank you!
Just following up now to see if you have any ideas about the calibration issue - whether it is a problem that my avatars are "levitating" and if so what to do about it.
Thanks very much!
Laura
Just following up now to see if you have any ideas about the calibration issue - whether it is a problem that my avatars are "levitating" and if so what to do about it.
Thanks very much!
Laura
-
Matt Petrucci
- Posts: 166
- Joined: Fri Feb 24, 2012 11:49 am
Re: Analysis and calibration problems
Hi Laura,
Great to hear the analysis problem got fixed!
One reason a participant may be levitating is there is an errant point in the kinematics where keypoint goes way below the surface of the floor (you can verify this if you load the TRC into OpenSim and visualize or plot the markers). This can often happen with occlusions or if a person is wearing black or darker colored pants where it's hard to see the joints. Or, if you have a very long capture volume and the person becomes very small at the end of the trial, it makes it hard to see smaller segments like the foot.
Ways to clear this up is to shorten your capture volume (or stop the trial earlier) and/or wear lighter colored pants (or shorts are even better to see the knees and angles). You could also add a third camera to help with occlusions.
Hope this helps,
Great to hear the analysis problem got fixed!
One reason a participant may be levitating is there is an errant point in the kinematics where keypoint goes way below the surface of the floor (you can verify this if you load the TRC into OpenSim and visualize or plot the markers). This can often happen with occlusions or if a person is wearing black or darker colored pants where it's hard to see the joints. Or, if you have a very long capture volume and the person becomes very small at the end of the trial, it makes it hard to see smaller segments like the foot.
Ways to clear this up is to shorten your capture volume (or stop the trial earlier) and/or wear lighter colored pants (or shorts are even better to see the knees and angles). You could also add a third camera to help with occlusions.
Hope this helps,