forums.naturalpoint.com

Hi,
I tried to perform a data collection with three markers. I loaded a rigid body previously prepared (constituted by three markers again) and I did the recording. The rigid body never "attached" to the three markers, probably because of a difference in the relative distances among the markers. However, the "trajectorize" and "convert to c3d" functions seemingly worked.

Usually, when the model is lost during a recording, in the c3d file I find "0" for the positions of the markers which constituted the lost model while the real coordinates are recorded under the name "Unlabeled".

The c3d files I got, instead, are corrupted: even though the number of seen markers is correct (e.g. 6 = 3 of the model lost + 3 of the actually seen markers) no coordinates are saved.

Unfortunately, I really need to have access to these data, as I probably won't be able to perform a new collection. Could someone suggest me a way to recover them?

Thank you in advance!

A few more details: I'm working with Arena 1.4 on Windows XP.
The .pt3 files are ok, as I can replay them in the software.

As a first solution,I'm just copying the 3d coordinates of the single markers that become visible when I click on each of them in the main Arena window during the replay

Of course any suggestion for a smarter solution (e.g. a .pt3 reader, or a different exporting procedure) would be appreciated!

Do you still have the recorded pt2 data?

Have you tried creating a new RB out of the recorded data?

Hi Vincent and sorry for the late answer, but I had only recently the chance to come back to the lab where Arena is installed.

I tried to build the RB, but it works for a single take only. To get all the c3d, I would have to create as many RB as the takes, and the process results even longer than writing the coordinates by hand.

A little explanation on why the problem occurred. I needed to record the position of two markers at the tip of the index and the thumb. To be able to build a sort of RB I added a third marker in between. On each trial the distance between thumb and index is different, so the RB cannot be attached to the markers for most of the takes. As the "trajectorize" function worked and the "export C3D" seemed to work I did not worry about this fact.

For the subsequent experiments I overcame the problem by creating the RB with the fingers being in contact and ended all takes asking the subjects to close their hand. Indeed, a single frame with the model attached seems enough to let the export work.

However I still need to recover the data of the first experiment. I have both the pt2 and the working pt3 data.

Thanks!

OK, if the distance, between the different takes, is different for the markers, then the RB you creat for one take, is not going to be able to auto assign, during a different.