scaling_preview static pose feature

Provide easy-to-use, extensible software for modeling, simulating, controlling, and analyzing the neuromusculoskeletal system.
POST REPLY
User avatar
Isotta Rigoni
Posts: 6
Joined: Thu May 10, 2018 8:31 am

scaling_preview static pose feature

Post by Isotta Rigoni » Tue Jun 05, 2018 9:24 am

Good afternoon everyone,
I am trying to scale my model according to the tutorial https://simtk-confluence.stanford.edu/d ... th+Scaling but I am not really understanding how to use the Preview Static Pose feature. From what I got, if we do not tick that option, the software will move the model markers to the experimental markers (which have already been scaled according to the SetUp_Scale.xml file) without showing the position error between these. On the other side, if we tick the Preview Static Pose we can check how close the model markers are to the experimental ones and in case, we can reposition the model markers in MarkerSt.xml and try again. Assuming that what I said is correct, could you tell me why the bones geometry looks like it has not been scaled when selecting the Preview Static Pose? In "Preview static pose.png", the geometry of the model is bigger than what it should be (and bigger than the model geometry in "NO preview static pose.png"). I don't know what I am not doing correctly. Do you have any suggestion?
Thanks all,
Isotta
Attachments
Preview static pose.PNG
Preview static pose.PNG (104.42 KiB) Viewed 504 times
NO preview static pose.PNG
NO preview static pose.PNG (121.58 KiB) Viewed 504 times
both.PNG
both.PNG (205.93 KiB) Viewed 504 times

Tags:

User avatar
jimmy d
Posts: 1375
Joined: Thu Oct 04, 2007 11:51 pm

Re: scaling_preview static pose feature

Post by jimmy d » Tue Jun 05, 2018 11:08 am

Hi, Isotta-

Unfortunately I am not getting the same behavior as you are seeing. Below is an image of the original model, a scaled model with Preview Static pose checked (Green), and a scaled model with Preview Static Pose left unchecked (red). The Green and Red models have had the same scaling factors applied to both.
Screen Shot 2018-06-05 at 10.21.42 AM.jpg
Screen Shot 2018-06-05 at 10.21.42 AM.jpg (46.62 KiB) Viewed 479 times

I also confirmed that they are the same by performing a difference check of the models in Notepad++. The Green and Red models only differ with the model marker placement.

Perhaps you aren't checking the Scaled Model section, so your posed model isn't having any scale factors applied to it? Below are my settings for scaling the model and previewing the static pose. Try and test by using the Gait2354 example files and see what you get.
Screen Shot 2018-06-05 at 10.16.44 AM.jpg
Screen Shot 2018-06-05 at 10.16.44 AM.jpg (68.63 KiB) Viewed 479 times

User avatar
Isotta Rigoni
Posts: 6
Joined: Thu May 10, 2018 8:31 am

Re: scaling_preview static pose feature

Post by Isotta Rigoni » Wed Jun 06, 2018 6:58 am

Hi James,
thank for your reply! I have tried again, but unfortunately some times I get your results and some others I get the unscaled model when checking Preview Static Pose. When this happens, I just close the software and reopen it again, but I don't really know why I am having this problem. If you have any ideas, let me know!
Can I bother you with another issue?
I scaled the model to a point I believe is good enough ("scaled_model.PNG") and run inverse kinematics with a dynamic trial ("gait.PNG" is a frame from it) and everything goes well. When I run inverse kinematics with a different trial (a trial from a whole body vibration treatment, with the subject keeping a squat position on a vibrating platform) I get something completely wrong (see "vibration_x.PNG" and "vibrations_y.PNG"). I don't know what this is caused by, but I thought it might be because the scaling process works better if the posture of the subject is similar to the one assumed during the dynamic trial. Does it make any sense? I therefore tried to redo the scaling using only one frame of the vibration trial (when the subject is keeping a stable squat position) and using the marker set I used for the previous trial. What I get is what you can see in "static_squat.PNG" and strangely the experimental markers do not match the model ones: why is that?
I hope you have any ideas :)
Thank you,
Isotta
Attachments
vibrations_x.PNG
vibrations_x.PNG (48.44 KiB) Viewed 451 times
vibration_y.PNG
vibration_y.PNG (28.21 KiB) Viewed 451 times
static_squat.PNG
static_squat.PNG (48.14 KiB) Viewed 451 times
gait.PNG
gait.PNG (40.86 KiB) Viewed 451 times
scaled_model.PNG
scaled_model.PNG (58.16 KiB) Viewed 451 times

POST REPLY