Method and System for Video Action Classification by Mixing 2D and 3D Features

    公开(公告)号:US20210357647A1

    公开(公告)日:2021-11-18

    申请号:US15931796

    申请日:2020-05-14

    Inventor: Han Na Rei Odaira

    Abstract: A method, system, and computer program product provide for video action classification by selecting a first video frame and a first plurality of video frames from a received video to process the first video frame with a 2D convolutional neural network processing pathway to extract spatial features classifying the first video frame, and to process the first plurality of video frames with a 3D convolutional neural network processing pathway to extract spatiotemporal features classifying the first plurality of video frames so that the spatial features are combined with the spatiotemporal features to generate a classification label for the video action.

    Approach to automated detection of dominant errors in cloud provisions

    公开(公告)号:US11231985B1

    公开(公告)日:2022-01-25

    申请号:US16935123

    申请日:2020-07-21

    Abstract: A system is configured to determine a dominant error causing a provisioning step to become stuck during provisioning of a machine in a cloud environment. The system includes memory for storing instructions, and a processor configured to execute said instructions to determine an inverse error frequency (IEF) value for pre-intervention errors in a set of intervention provisioning data; determine a dominant error for a provision during said provisioning step in said set of intervention provisioning data based on a pre-intervention error that has a maximum IEF value; determine a duration frequency (DuF) value for the provision at said provisioning step for provisions in a set of non-intervention provisioning data; and determine said dominant error for each provision during said provisioning step in said set of non-intervention provisioning data based on an error that resulted in DuF value.

    Method and system for video action classification by mixing 2D and 3D features

    公开(公告)号:US11276249B2

    公开(公告)日:2022-03-15

    申请号:US15931796

    申请日:2020-05-14

    Inventor: Han Na Rei Odaira

    Abstract: A method, system, and computer program product provide for video action classification by selecting a first video frame and a first plurality of video frames from a received video to process the first video frame with a 2D convolutional neural network processing pathway to extract spatial features classifying the first video frame, and to process the first plurality of video frames with a 3D convolutional neural network processing pathway to extract spatiotemporal features classifying the first plurality of video frames so that the spatial features are combined with the spatiotemporal features to generate a classification label for the video action.

    APPROACH TO AUTOMATED DETECTION OF DOMINANT ERRORS IN CLOUD PROVISIONS

    公开(公告)号:US20220027222A1

    公开(公告)日:2022-01-27

    申请号:US16935123

    申请日:2020-07-21

    Abstract: A system is configured to determine a dominant error causing a provisioning step to become stuck during provisioning of a machine in a cloud environment. The system includes memory for storing instructions, and a processor configured to execute said instructions to determine an inverse error frequency (IEF) value for pre-intervention errors in a set of intervention provisioning data; determine a dominant error for a provision during said provisioning step in said set of intervention provisioning data based on a pre-intervention error that has a maximum IEF value; determine a duration frequency (DuF) value for the provision at said provisioning step for provisions in a set of non-intervention provisioning data; and determine said dominant error for each provision during said provisioning step in said set of non-intervention provisioning data based on an error that resulted in DuF value.

Patent Agency Ranking