公开(公告)号：US20230035282A1

公开(公告)日：2023-02-02

申请号：US17384522

申请日：2021-07-23

Applicant: QUALCOMM Incorporated

Inventor： Kuang-Man HUANG ,  Min-Hui Lin ,  Ke-Li CHENG ,  Michel Adib SARKIS

IPC: G06N3/08 ,  G06K9/00 ,  G06T17/00

Abstract: Systems and techniques are provided for generating one or more models. For example, a process can include obtaining a plurality of input images corresponding to faces of one or more people during a training interval. The process can include determining a value of the coefficient representing at least the portion of the facial expression for each of the plurality of input images during the training interval. The process can include determining, from the determined values of the coefficient representing at least the portion of the facial expression for each of the plurality of input images during the training interval, an extremum value of the coefficient representing at least the portion of the facial expression during the training interval. The process can include generating an updated bounding value for the coefficient representing at least the portion of the facial expression based on the initial bounding value and the extremum value.

公开(公告)号：US11977979B2

公开(公告)日：2024-05-07

申请号：US17384522

申请日：2021-07-23

Applicant: QUALCOMM Incorporated

Inventor： Kuang-Man Huang ,  Min-Hui Lin ,  Ke-Li Cheng ,  Michel Adib Sarkis

IPC: G06V40/16 ,  G06N3/08 ,  G06T17/00

CPC classification number: G06N3/08 ,  G06T17/00 ,  G06V40/174 ,  G06T2210/44

Abstract: Systems and techniques are provided for generating one or more models. For example, a process can include obtaining a plurality of input images corresponding to faces of one or more people during a training interval. The process can include determining a value of the coefficient representing at least the portion of the facial expression for each of the plurality of input images during the training interval. The process can include determining, from the determined values of the coefficient representing at least the portion of the facial expression for each of the plurality of input images during the training interval, an extremum value of the coefficient representing at least the portion of the facial expression during the training interval. The process can include generating an updated bounding value for the coefficient representing at least the portion of the facial expression based on the initial bounding value and the extremum value.