Closed stowerslab closed 1 year ago
I think the fact that the video is flipped upside down here is just a quirk of ipywidgets. Thanks for pointing this out!
But you do raise a more substantive point, which is that the heading-angle component of the modeling framework may not be appropriate for head-fixed setups. There is actually a setting in the config called "fix_heading" that addresses this. In your case you could try setting "fix_heading=True". This is still an experimental feature and in general we haven't applied keypoint-MoSeq to headfixed data before. It might be worth trying both options (fix_heading=True/False) and seeing which one produces more reasonable syllables.
hey caleb :) tried both True/False for the heading and it didnt change the orientation (still upside down)
The fix_heading
parameter isn't related to the frames being upside down during calibration.
fix_heading
parameter determines whether points are rotationally aligned during model fitting. gotcha!
We have videos of headfixed mice that are being flipped upside down during the error calibration and PCA extraction (probably because of rotational alignment) so we are wondering if that can be made optional or if the direction of alignment can be changed