classify behaviors with 3D pose estimation

YttriLab / B-SOID

Behavioral segmentation of open field in DeepLabCut, or B-SOID ("B-side"), is a pipeline that pairs unsupervised pattern recognition with supervised classification to achieve fast predictions of behaviors that are not predefined by users.

GNU General Public License v3.0

191 stars 54 forks source link

classify behaviors with 3D pose estimation #34

Open lifengyeh opened 3 years ago

lifengyeh commented 3 years ago

I have tried the codes to analyze my deeplabcut data. The result is amazing! Now I am tracking animals with one camera and generating 2D trajectories; but planning to use 2 camera for 3D pose estimation. The B-SOID seems designed for 2D tracking. Would it have future extension for 3D trajectories? Thanks!

runninghsus commented 3 years ago

Hi @lifengyeh

we have a working code for 3D (x,y,z). We plan on making sure that it is integrated flawlessly with the current solution before releasing. If you want the 3D code, you can contact me directly on slack