知搜-全球专利检索

  1. Login
  2. Sign Up

Search Results

15 records (took 0.029 seconds)

    FEATURE-BASED VIDEO ANNOTATION
    2.
    发明申请
    FEATURE-BASED VIDEO ANNOTATION 有权

    公开(公告)号：US20220207873A1

    公开(公告)日：2022-06-30

    申请号：US17548859

    申请日：2021-12-13

    Applicant: Google LLC

    Inventor： Balakrishnan Varadarajan ,  George Dan Toderici ,  Apostol Natsev ,  Nitin Khandelwal ,  Sudheendra Vijayanarasimhan ,  Weilong Yang ,  Sanketh Shetty

    IPC: G06V20/40 ,  G06F16/783 ,  G06F16/78 ,  G06V10/42 ,  G06K9/62 ,  H04N5/265

    Abstract: A system and methodology provide for annotating videos with entities and associated probabilities of existence of the entities within video frames. A computer-implemented method identifies an entity from a plurality of entities identifying characteristics of video items. The computer-implemented method selects a set of features correlated with the entity based on a value of a feature of a plurality of features, determines a classifier for the entity using the set of features, and determines an aggregation calibration function for the entity based on the set of features. The computer-implemented method selects a video frame from a video item, where the video frame having associated features, and determines a probability of existence of the entity based on the associated features using the classifier and the aggregation calibration function.

    Framework for training machine-learned models on extremely large datasets
    5.
    发明授权
    Framework for training machine-learned models on extremely large datasets 有权

    公开(公告)号：US11295171B2

    公开(公告)日：2022-04-05

    申请号：US16657042

    申请日：2019-10-18

    Applicant: Google LLC

    Inventor： Joonseok Lee ,  Balakrishnan Varadarajan ,  Ariel Gordon ,  Apostol Ivanov Natsev ,  Seong Jae Hwang

    IPC: G06K9/62 ,  G06N20/00 ,  G06K9/00 ,  G06K1/00

    Abstract: A MapReduce-based training framework exploits both data parallelism and model parallelism to scale training of complex models. Particular model architectures facilitate and benefit from use of such training framework. As one example, a machine-learned model can include a shared feature extraction portion configured to receive and process a data input to produce an intermediate feature representation and a plurality of prediction heads that are configured to receive and process the intermediate feature representation to respectively produce a plurality of predictions. For example, the data input can be a video and the plurality of predictions can be a plurality of classifications for content of the video (e.g., relative to a plurality of classes).

    Discovery of news-related content
    6.
    发明授权
    Discovery of news-related content 有权

    公开(公告)号：US10235428B2

    公开(公告)日：2019-03-19

    申请号：US15195105

    申请日：2016-06-28

    Applicant: Google LLC

    Inventor： Balakrishnan Varadarajan ,  Sudheendra Vijayanarasimhan ,  Sanketh Shetty ,  Nisarg Dilipkumar Kothari ,  Nicholas Delmonico Rizzolo

    IPC: G06F17/30 ,  G10L25/54 ,  H04L12/58 ,  H04L29/08

    Abstract: Techniques identify time-sensitive content and present the time-sensitive content to communication devices of users interested or potentially interested in the time-sensitive content. A content management component analyzes video or audio content, and extracts information from the content and determines whether the content is time-sensitive content, such as recent news-related content, based on analysis of the content and extracted information. The content management component evaluates user-related information and the extracted information, and determines whether a user(s) is likely to be interested in the time-sensitive content based on the evaluation results. The content management component sends a notification to the communication device(s) of the user(s) in response to determining the user(s) is likely to be interested in the time-sensitive content.

    Framework for Training Machine-Learned Models on Extremely Large Datasets
    9.
    发明申请
    Framework for Training Machine-Learned Models on Extremely Large Datasets 有权

    公开(公告)号：US20210117728A1

    公开(公告)日：2021-04-22

    申请号：US16657042

    申请日：2019-10-18

    Applicant: Google LLC

    Inventor： Joonseok Lee ,  Balakrishnan Varadarajan ,  Ariel Gordon ,  Apostol Ivanov Natsev ,  Seong Jae Hwang

    IPC: G06K9/62 ,  G06K9/00 ,  G06N20/00

    Abstract: A MapReduce-based training framework exploits both data parallelism and model parallelism to scale training of complex models. Particular model architectures facilitate and benefit from use of such training framework. As one example, a machine-learned model can include a shared feature extraction portion configured to receive and process a data input to produce an intermediate feature representation and a plurality of prediction heads that are configured to receive and process the intermediate feature representation to respectively produce a plurality of predictions. For example, the data input can be a video and the plurality of predictions can be a plurality of classifications for content of the video (e.g., relative to a plurality of classes).

    FEATURE-BASED VIDEO ANNOTATION
    10.
    发明申请
    FEATURE-BASED VIDEO ANNOTATION 审中-公开

    公开(公告)号：US20200082173A1

    公开(公告)日：2020-03-12

    申请号：US16687118

    申请日：2019-11-18

    Applicant: Google LLC

    Inventor： Balakrishnan Varadarajan ,  George Dan Toderici ,  Apostol Natsev ,  Nitin Khandelwal ,  Sudheendra Vijayanarasimhan ,  Weilong Yang ,  Sanketh Shetty

    IPC: G06K9/00 ,  H04N5/265 ,  G06K9/62 ,  G06K9/52 ,  G06F16/78 ,  G06F16/783

    Abstract: A system and methodology provide for annotating videos with entities and associated probabilities of existence of the entities within video frames. A computer-implemented method identifies an entity from a plurality of entities identifying characteristics of video items. The computer-implemented method selects a set of features correlated with the entity based on a value of a feature of a plurality of features, determines a classifier for the entity using the set of features, and determines an aggregation calibration function for the entity based on the set of features. The computer-implemented method selects a video frame from a video item, where the video frame having associated features, and determines a probability of existence of the entity based on the associated features using the classifier and the aggregation calibration function.

Patent Agency Ranking