CSAILVision/GazeCapture

Eye Bounding Box

bocsiboti opened this issue · 1 comments

Hi,
As far as I know the iOS face/eye detection service does not provide bounding boxes for eyes, only eye and eyebrow control points (landmarks). What was the procedure/algorithm that you used to generate bounding boxes for eyes from the provided information?
Thanks, Botond

It was a heuristic based on face rectangle size and eye center positions. The eye rectangle is centered around the eye center and has a size proportional to the face size.