Face rigid body model and fixation point detection method and device and storage medium

Abstract

The invention discloses a face rigid body model, a fixation point detection method and device and a storage medium, and the method comprises the steps: obtaining a first visible light face image of a current user, which is forwardly shot by an image sensor, and obtaining a preset number of feature points of the face of the current user from the first visible light face image; determining a left eyeball center three-dimensional coordinate, a right eyeball center three-dimensional coordinate and a face center three-dimensional coordinate of the current user based on the preset number of feature points; and respectively establishing a left eye vector and a right eye vector from the left eyeball center three-dimensional coordinate and the right eyeball center three-dimensional coordinate to the face center three-dimensional coordinate, and mapping the left eye vector and the right eye vector to the face three-dimensional model to obtain a face rigid body model of the current user.

Description

technical field [0001] The present invention relates to the field of machine vision, in particular to a face rigid body model and a method, device and storage medium for gaze point detection. Background technique [0002] At present, the spatial gaze tracking system mostly adopts the method of polynomial model mapping. Generally, the 9-point calibration method is adopted, that is, the human eye looks at 9 calibration points on the screen (equivalent to a two-dimensional plane) in turn, so as to obtain the distance between the pupil and the screen. The coordinate mapping model of , which is used to calculate the coordinates of the gaze point on the screen. [0003] Although the 9-point calibration method has high accuracy when the face remains stationary, when the face moves or rotates, the polynomial model obtained from the historical 9-point calibration is no longer suitable for The gaze calculation under the current face pose makes the calculation accuracy of the gaze poi...

AI Technical Summary

Problems solved by technology

[0003] Although the 9-point calibration method has high accuracy when the face remains still, when the face moves or rotates, the polynomial model obtained by the historical 9-point calibration is no longer applicable due to the change of the face pose. The line of sight calculation under the current face pose makes the calculation accuracy of the gaze point drop significantly

Images