METHODS AND SYSTEM FOR TRAINING AND REINFORCING COMPUTER VISION MODELS USING DISTRIBUTED COMPUTING

    公开(公告)号:US20220038534A1

    公开(公告)日:2022-02-03

    申请号:US16944737

    申请日:2020-07-31

    Abstract: Systems and methods described herein perform computer vision (CV) model training in a distributed edge network. Regional Multi-access Edge Compute (MEC) clusters are provided with a local copy of a CV model and a local synthetic training data generator. A MEC cluster receives client data requiring computer vision and applies the local copy of the CV model to the client data. The MEC cluster detects an exception to the local copy of the CV model and generates local synthetic training data for the exception. The MEC cluster updates, based on the local synthetic training data, the local copy of the CV model to form an updated local CV model. The MEC cluster sends the local synthetic training data and the updated local CV model to a central network. The central network uses the local synthetic training data to update a master CV model and any other interconnected CV models.

    Methods and system for training and reinforcing computer vision models using distributed computing

    公开(公告)号:US11463517B2

    公开(公告)日:2022-10-04

    申请号:US16944737

    申请日:2020-07-31

    Abstract: Systems and methods described herein perform computer vision (CV) model training in a distributed edge network. Regional Multi-access Edge Compute (MEC) clusters are provided with a local copy of a CV model and a local synthetic training data generator. A MEC cluster receives client data requiring computer vision and applies the local copy of the CV model to the client data. The MEC cluster detects an exception to the local copy of the CV model and generates local synthetic training data for the exception. The MEC cluster updates, based on the local synthetic training data, the local copy of the CV model to form an updated local CV model. The MEC cluster sends the local synthetic training data and the updated local CV model to a central network. The central network uses the local synthetic training data to update a master CV model and any other interconnected CV models.

    SYSTEMS AND METHODS FOR DYNAMIC AUDIOVISUAL CONFERENCING IN VARYING NETWORK CONDITIONS

    公开(公告)号:US20230318864A1

    公开(公告)日:2023-10-05

    申请号:US18333593

    申请日:2023-06-13

    CPC classification number: H04L12/1827 H04M3/568 H04M3/567 H04L12/1831

    Abstract: Embodiments described herein provide for the dynamic adjustment of traffic associated with audiovisual conferences or other types of communication sessions in situations where a network connection of one or more conference participants exhibits issues that may affect audio and/or video quality. The adjustment may include the determination at a particular conference participant (e.g., a particular User Equipment (“UE”)) of degraded network conditions, the generation of condensed conference information at the UE, and the outputting of the condensed conference information via the network. The condensed conference information may be included in control signaling. The condensed conference information may be used to generate reconstructed conference information, which may be distributed to other conference participants.

    Systems and methods for dynamic audiovisual conferencing in varying network conditions

    公开(公告)号:US11716214B2

    公开(公告)日:2023-08-01

    申请号:US17378984

    申请日:2021-07-19

    CPC classification number: H04L12/1827 H04L12/1831 H04M3/567 H04M3/568

    Abstract: Embodiments described herein provide for the dynamic adjustment of traffic associated with audiovisual conferences or other types of communication sessions in situations where a network connection of one or more conference participants exhibits issues that may affect audio and/or video quality. The adjustment may include the determination at a particular conference participant (e.g., a particular User Equipment (“UE”)) of degraded network conditions, the generation of condensed conference information at the UE, and the outputting of the condensed conference information via the network. The condensed conference information may be included in control signaling. The condensed conference information may be used to generate reconstructed conference information, which may be distributed to other conference participants.

    SYSTEMS AND METHODS FOR DYNAMIC AUDIOVISUAL CONFERENCING IN VARYING NETWORK CONDITIONS

    公开(公告)号:US20230013881A1

    公开(公告)日:2023-01-19

    申请号:US17378984

    申请日:2021-07-19

    Abstract: Embodiments described herein provide for the dynamic adjustment of traffic associated with audiovisual conferences or other types of communication sessions in situations where a network connection of one or more conference participants exhibits issues that may affect audio and/or video quality. The adjustment may include the determination at a particular conference participant (e.g., a particular User Equipment (“UE”)) of degraded network conditions, the generation of condensed conference information at the UE, and the outputting of the condensed conference information via the network. The condensed conference information may be included in control signaling. The condensed conference information may be used to generate reconstructed conference information, which may be distributed to other conference participants.

Patent Agency Ranking