Invention Grant
- Patent Title: Facial anti-spoofing method using variances in image properties
-
Application No.: US16649571Application Date: 2018-07-12
-
Publication No.: US11314966B2Publication Date: 2022-04-26
- Inventor: John F. Sheets , Sunpreet Singh Arora , Lacey Best-Rowden , Kim R. Wagner
- Applicant: Visa International Service Association
- Applicant Address: US CA San Francisco
- Assignee: Visa International Service Association
- Current Assignee: Visa International Service Association
- Current Assignee Address: US CA San Francisco
- Agency: Kilpatrick Townsend & Stockton LLP
- International Application: PCT/US2018/041795 WO 20180712
- International Announcement: WO2019/060023 WO 20190328
- Main IPC: G06K9/00
- IPC: G06K9/00 ; H04N5/232 ; G06F21/32 ; G06F21/45 ; G06K9/46

Abstract:
Embodiments of the invention are directed to a system and methods for determining a likelihood that an image that includes a user is a spoof or fake. In some embodiments, focus values are determined for various parts of the image during a focus sweep. The focus values may represent a focus point at which a particular part of the image is sharp. In some embodiments, the focus values may be determined only for the sections of the image that correspond to a face within the image. From the focus values, the system may determine a relative depth of various parts of the image. Using the relative depths, the system may generate a rough depth map for the image. The depth map may be analyzed to determine a likelihood that the image is authentic.
Public/Granted literature
- US20200218886A1 Facial Anti-Spoofing Method Using Variances In Image Properties Public/Granted day:2020-07-09
Information query