DeepLabCut
Refer to API:
neuralib.deeplabcut
from neuralib.deeplabcut import read_dlc
# load DeepLabCut output from .h5 OR .csv and its metadata (.pkl)
dlc_df = read_dlc("/path/to/output.csv", meta_file="/path/to/meta.pkl")
# access dataframe with all joints
df = dlc_df.dataframe()
# list available joints
print("Joints:", dlc_df.joints)
# access metadata
print("Frames per second:", dlc_df.fps)
print("Total frames:", dlc_df.nframes)
# access a specific joint's data
nose_df = dlc_df.get_joint("Nose").dataframe()
print(nose_df)