MouseLand / facemap

Framework for predicting neural activity from mouse orofacial movements tracked using a pose estimation model. Package also includes singular value decomposition (SVD) of behavioral videos.
https://facemap.readthedocs.io
GNU General Public License v3.0
157 stars 61 forks source link

Pupil SVD Use Case for Just Pupil Data #148

Closed KemalOz21 closed 4 months ago

KemalOz21 commented 5 months ago

To whom it may concern, First of all, congratulations on publishing and developing Facemap AI. The paper was intriguing for me to read. I currently have mouse pupillometry data and I want to use Facemap's pupil SVD option to find the spatially and temporally changing pupil area. Can the program work for just a pupil video, or does it need to have the whole mouse's face as shown in the pupil SVD documentation?

Thanks

carsen-stringer commented 5 months ago

Hi @KemalOz21 , you can track the pupil with facemap without a full video -- you will want to draw the ROI for the pupil just around the eye anyway. This video is a bit out-dated but the ROIs work the same way in the latest version: https://www.youtube.com/watch?v=Rq8fEQ-DOm4

KemalOz21 commented 5 months ago

Thank you so much for sharing the video. It was very helpful