公开(公告)号：US20200382745A1

公开(公告)日：2020-12-03

申请号：US16689458

申请日：2019-11-20

Applicant: Apple Inc.

Inventor： Johnny Trenh ,  Hsi-Jung Wu ,  Sarah K. Herrlinger ,  Xiaoxia Sun ,  Ian J. Baird ,  Dazhong Zhang ,  Xiaosong Zhou ,  Christopher M. Garrido

IPC: H04N7/14 ,  H04N7/15 ,  G06N3/08 ,  G06F3/01

Abstract: Techniques are presented for managing for visual prominence of participants in a video conference, including conferences where participants communicate visually, such as with sign language. According to these techniques, a visual prominence indication of a participant in a video conference may be estimated, a video stream of the participant may be encoded, and the encoded video stream may be transmitted along with an indication of the estimated visual prominence to a receiving device in the video conference.

公开(公告)号：US11847823B2

公开(公告)日：2023-12-19

申请号：US17339249

申请日：2021-06-04

Applicant: Apple Inc.

Inventor： Xiaoxia Sun ,  Jiefu Zhai ,  Ke Zhang ,  Xiaosong Zhou ,  Hsi-Jung Wu

IPC: G06V10/82 ,  G06T3/40 ,  G06V40/20 ,  G06V20/40 ,  G06N3/045 ,  G06V10/25 ,  G06V20/20 ,  G06V40/10

CPC classification number: G06V10/82 ,  G06N3/045 ,  G06T3/40 ,  G06V10/25 ,  G06V20/20 ,  G06V20/46 ,  G06V40/113 ,  G06V40/28

Abstract: Video object and keypoint location detection techniques are presented. The system includes a detection system for generation locations of an object's keypoints along with probabilities associated with the locations, and a stability system for stabilizing keypoint locations of the detected objects. In some aspects, the generated probabilities are two-dimensional array correspond locations within input images, and stability system fits the generated probabilities to a two-dimensional probability distribution function.

公开(公告)号：US20230147442A1

公开(公告)日：2023-05-11

申请号：US17831738

申请日：2022-06-03

Applicant: Apple Inc.

Inventor： Shujie Liu ,  Jiefu Zhai ,  Xiaosong Zhou ,  Hsi-Jung Wu ,  Ke Zhang ,  Xiaoxia Sun ,  Jian Li

IPC: G06N3/045

CPC classification number: G06N3/045

Abstract: In an example method, a system accesses first input data and a machine learning architecture. The machine learning architecture includes a first module having a first neural network, a second module having a second neural network, and a third module having a third neural network. The system generates a first feature set representing a first portion of the first input data using the first neural network, and a second feature set representing a second portion of the first input data using the second neural network. The system generates, using the third neural network, first output data based on the first feature set and the second feature set.

公开(公告)号：US11165989B2

公开(公告)日：2021-11-02

申请号：US16689458

申请日：2019-11-20

Applicant: Apple Inc.

Inventor： Johnny Trenh ,  Hsi-Jung Wu ,  Sarah K. Herrlinger ,  Xiaoxia Sun ,  Ian J. Baird ,  Dazhong Zhang ,  Xiaosong Zhou ,  Christopher M. Garrido

IPC: H04N7/14 ,  G06F3/01 ,  G06N3/08 ,  H04N7/15

Abstract: Techniques are presented for managing for visual prominence of participants in a video conference, including conferences where participants communicate visually, such as with sign language. According to these techniques, a visual prominence indication of a participant in a video conference may be estimated, a video stream of the participant may be encoded, and the encoded video stream may be transmitted along with an indication of the estimated visual prominence to a receiving device in the video conference.