Adding New Markers to a .trc

[Aaron Godfrey]

I have a program I'm using to compare marker position over time, but I need to add some virtual markers to it to complete my analysis. This might be a dumb question, but is there anyway to re-calculate the trace with virtual markers and export those results? There's plenty of options for saving .mots and whatnot, but I haven't found a way to do traces - I don't want joint angles, I want marker positions. I feel like I'm missing something obvious...
Any help is appreciated. Thanks!

[Thomas Uchida]

Not sure what you mean by "trace", but the IK Tool can output marker trajectories over time; see viewtopicPhpbb.php?f=91&t=8311.

[Aaron Godfrey]

I was referring to a .trc file, which is what I use to run IK.

The tool option you pointed out might work, though. Thanks!

[Aaron Godfrey]

Having looked into this a little more, it isn't the solution I was looking for. While it does put out an STO of marker locations, it does not put out the location of the virtual markers (i.e., markers that I add to the model prior to running IK, that are tied to various bodies - for example, I'm adding a new marker at the elbow and attaching it to the humerus). Is there a way I can get the output of the marker motion for that new marker?

[Thomas Uchida]

it does not put out the location of the virtual markers (i.e., markers that I add to the model prior to running IK, that are tied to various bodies - for example, I'm adding a new marker at the elbow and attaching it to the humerus). Is there a way I can get the output of the marker motion for that new marker?

I haven't dug into the code, but perhaps the Tool reports only the markers that are being tracked (i.e., on which the objective function depends). If so, a quick fix might be to add columns for the virtual markers in your TRC file (with junk trajectories) and set their tracking weights to zero.

[Aaron Godfrey]

That did it. Thanks again!

[Thomas Uchida]

Great; thanks for reporting back. You might also want to investigate the scripting interface (https://simtk-confluence.stanford.edu:8 ... /Scripting), which would be a more elegant and flexible approach.