Invention Grant
- Patent Title: Determining quality of images for user identification
-
Application No.: US15716651Application Date: 2017-09-27
-
Publication No.: US10803297B2Publication Date: 2020-10-13
- Inventor: Yuk L. Chan , Deepti M. Naphade , Tin Hang To
- Applicant: International Business Machines Corporation
- Applicant Address: US NY Armonk
- Assignee: International Business Machines Corporation
- Current Assignee: International Business Machines Corporation
- Current Assignee Address: US NY Armonk
- Agent Jose A. Medina-Cruz; William H. Hartwell
- Main IPC: G06K9/00
- IPC: G06K9/00 ; H04L29/06 ; G06F16/583 ; G06F16/58

Abstract:
In an approach to determining quality score of images for user identification, one or more processors receive a set of input images for a first user. One or more processors may determine a quality score for each image in the set of input images with respect to an ability of each respective image to identify the first user, wherein the quality score pertains to both: (i) a quality the image, and (ii) a quality of a subject of the image. One or more processors may additionally determine a first set of output images based on the quality score for each image in the set of input images. One or more processors may further identify the first user based on the first set of output images.
Public/Granted literature
- US20190095702A1 DETERMINING QUALITY OF IMAGES FOR USER IDENTIFICATION Public/Granted day:2019-03-28
Information query