Title:
MACHINE-LEARNING-BASED EYE TRACKING DEVICE AND METHOD
Document Type and Number:
WIPO Patent Application WO/2022/215952
Kind Code:
A1
Abstract:
The present invention relates to a machine-learning-based eye tracking device and method. A machine-learning-based eye tracking device according to one embodiment of the present invention comprises: an input unit for inputting an image including a face; a feature point detection unit for detecting a first feature point from the image including the face; a face direction detection unit for detecting face direction on the basis of the detected first feature point; an eyeball direction detection unit for detecting eyeball direction, which is a characteristic of the eyeball, from the detected first feature point; a model training unit for training an eye tracking model by using the detected feature point and directions as inputs; and an eye tracking unit for performing eye tracking by using the trained eye tracking model.
Inventors:
NOH YOO HUN (KR)
Application Number:
PCT/KR2022/004627
Publication Date:
October 13, 2022
Filing Date:
April 04, 2022
Export Citation:
Assignee:
EMOCOG CO LTD (KR)
International Classes:
G06V40/18; G06N20/00; G06V10/40; G06V40/16
Domestic Patent References:
WO2020231401A1 | 2020-11-19 |
Foreign References:
KR20190040797A | 2019-04-19 | |||
KR101819164B1 | 2018-01-17 | |||
KR20150117553A | 2015-10-20 | |||
JP2007265367A | 2007-10-11 | |||
JP6745518B1 | 2020-08-26 | |||
KR101094766B1 | 2011-12-20 |
Attorney, Agent or Firm:
THEWAVE IP LAW FIRM (KR)
Download PDF: