US 12,299,099 B2
Method and apparatus for continuous authentication
Myungsu Chae, Daejeon (KR)
Assigned to NOTA, INC., Daejeon (KR)
Filed by NOTA, INC., Daejeon (KR)
Filed on Jun. 17, 2022, as Appl. No. 17/843,520.
Application 17/843,520 is a continuation of application No. PCT/KR2020/001076, filed on Jan. 22, 2020.
Claims priority of application No. 10-2019-0170112 (KR), filed on Dec. 18, 2019; and application No. 10-2020-0008275 (KR), filed on Jan. 22, 2020.
Prior Publication US 2022/0318358 A1, Oct. 6, 2022
Int. Cl. G06F 21/32 (2013.01); G06T 7/246 (2017.01); G06T 7/70 (2017.01); G06V 40/16 (2022.01)
CPC G06F 21/32 (2013.01) [G06T 7/248 (2017.01); G06T 7/70 (2017.01); G06V 40/172 (2022.01); G06T 2207/10016 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/30201 (2013.01)] 14 Claims
OG exemplary drawing
 
1. An authentication method executed in a computer device, wherein the computer device comprises at least one processor configured to execute computer-readable instructions included in a memory, and
wherein the authentication method comprises:
receiving, by the at least one processor, image frames taken by a camera in succession;
detecting, by the at least one processor, a face area in the image frames;
tracking, by the at least one processor, a change in a location of the detected face area in the image frames; and
performing, by the at least one processor, continuous user authentication for the face area according to the change in the location by using the face area whose change in the location has been tracked and a deep learning model;
wherein the authentication method further comprises:
collecting, by the at least one processor, the image frames taken by the camera, the detected face area, and results of the user authentication; and
retraining, by the at least one processor, the deep learning model by using the collected image frames, the collected face area, and the collected results of the user authentication; and
wherein retraining the deep learning model comprises:
classifying the collected image frames, the collected face area into data successful in the user authentication and data failed in the user authentication based on the collected results of the user authentication;
training the deep learning model by using, as correct answer data, image frames and a corresponding face area successful in the user authentication along with user ID information; and
training the deep learning model by using, as incorrect data, image frames and a corresponding face area failed in the user authentication.