icatcherplus / icatcher_plus

iCatcher+: Robust and automated annotation of infant gaze from videos collected in laboratory, field, and online studies
GNU General Public License v3.0
17 stars 17 forks source link

Update how lowest face is chosen #40

Closed samwinebrake closed 1 year ago

samwinebrake commented 1 year ago

The previous lowest face selection (bbox = min(bboxes, key=lambda x: x[3] - x[1])) actually wasn't always selecting the lowest face ( because (0,0) coordinate is in the top left corner ). This is fixed, as well as a height/width ratio closer to 1 being favored. If you don't want this to be the default, I can change the addition of the ratio weight to a flag!