#$%^&*AU2018264135A120181206.pdf#####- 34 ABSTRACT Embodiments of the invention relate to a method of extracting eye velocity information from a video footage having a plurality of frames, comprising detecting at least part of an eye in at least two frames of the video footage, applying an optical flow algorithm to the at 5 least two frames of the video footage to extract pixel velocity information, and determining a statistical measure from the pixel velocity information within the detected at least part of the eye. Other embodiments of the invention relate to a method of extracting head image trajectory 10 information from a video footage having a plurality of frames, comprising detecting at least part of a facial region of the head image in at least two frames of the video footage, determining a measure of the movement of the at least part of a facial region between the at least two frames, and determining a transformation map from the measure of the movement. 153/11 Input Images 1(t) and l(t+1) 10 20 Limbal Edge Optic Flow Field Detection from 30 on (t) l(t+1) and I(t) Limbal image Velocity field Masked Velocity 4 Field 4 Mean Limbal FIGURE 3 Velocity (a) (b) Moving Limbus Model Simulated Response =v =.1 px-Jframe -V: I 7..ta E FIGURE 0.2- U)0 0.1. -0 -10 0 10 20 -W 0-i 0 10 20 FIGURE 4
