-
公开(公告)号：US20230267302A1
公开(公告)日：2023-08-24
申请号：US17940568
申请日：2022-09-08
Applicant: Google LLC
Inventor： Bryan Thomas Perozzi , Anton Tsitsulin , John Joseph Palowitch , Brandon Mayer
Abstract: Systems and methods for graph model search and/or for architecture insight can include training and testing a plurality of graph models. For example, the systems and methods can generate a plurality of synthetic graph datasets, which can then be utilized to train a plurality of graph models with varying graph model architectures. The trained graph models can then be evaluated based on outputs generated by the models based on test inputs. The evaluation data can then be utilized for providing particular graph model insight and/or may be utilized to enable task-specific graph model search.
-
公开(公告)号：US20220075995A1
公开(公告)日：2022-03-10
申请号：US16624450
申请日：2019-05-20
Applicant: Google LLC
Inventor： Brandon Mayer , John Finley
Abstract: Systems and methods for providing a multi-step identification process using a hybrid edge/server infrastructure to generate machine learned inferences are provided. One example method includes obtaining sensor data of an environment including one or more user. The method includes selecting at least a portion of the sensor data that is associated with the one or more users. The method includes generating, using one or more machine learning models, an intermediate representation of the selected portion of the sensor data that represents the face of the one or more users. The method includes transmitting the intermediate representation and a request for a user identification based on the intermediate representation, the remote computing system configured to perform additional steps of the multi-step identification process using the intermediate representation and the machine-learned system. The method includes receiving a user identifier for the one or more users.
-
Patent Agency Ranking
公开(公告)号：US12051270B2
公开(公告)日：2024-07-30
申请号：US17950727
申请日：2022-09-22
Applicant: Google LLC
Inventor： Brandon Mayer , John Finley
IPC: G06V40/16 , G06F21/32 , G06T7/70 , H04L65/403
CPC classification number: G06V40/172 , G06F21/32 , G06T7/70 , H04L65/403 , G06T2207/10016
Abstract: Systems and methods for providing a multi-step identification process using a hybrid edge/server infrastructure to generate machine learned inferences are provided. One example method includes obtaining sensor data of an environment including one or more user. The method includes selecting at least a portion of the sensor data that is associated with the one or more users. The method includes generating, using one or more machine learning models, an intermediate representation of the selected portion of the sensor data that represents the face of the one or more users. The method includes transmitting the intermediate representation and a request for a user identification based on the intermediate representation, the remote computing system configured to perform additional steps of the multi-step identification process using the intermediate representation and the machine-learned system. The method includes receiving a user identifier for the one or more users.
-
公开(公告)号：US20230092618A1
公开(公告)日：2023-03-23
申请号：US17950727
申请日：2022-09-22
Applicant: Google LLC
Inventor： Brandon Mayer , John Finley
IPC: G06V40/16 , G06T7/70 , G06F21/32 , H04L65/403
Abstract: Systems and methods for providing a multi-step identification process using a hybrid edge/server infrastructure to generate machine learned inferences are provided. One example method includes obtaining sensor data of an environment including one or more user. The method includes selecting at least a portion of the sensor data that is associated with the one or more users. The method includes generating, using one or more machine learning models, an intermediate representation of the selected portion of the sensor data that represents the face of the one or more users. The method includes transmitting the intermediate representation and a request for a user identification based on the intermediate representation, the remote computing system configured to perform additional steps of the multi-step identification process using the intermediate representation and the machine-learned system. The method includes receiving a user identifier for the one or more users.
-
公开(公告)号：US11468708B2
公开(公告)日：2022-10-11
申请号：US16624450
申请日：2019-05-20
Applicant: Google LLC
Inventor： Brandon Mayer , John Finley
IPC: H04L65/403 , G06V40/16 , G06T7/70 , G06F21/32
Abstract: Systems and methods for providing a multi-step identification process using a hybrid edge/server infrastructure to generate machine learned inferences are provided. One example method includes obtaining sensor data of an environment including one or more user. The method includes selecting at least a portion of the sensor data that is associated with the one or more users. The method includes generating, using one or more machine learning models, an intermediate representation of the selected portion of the sensor data that represents the face of the one or more users. The method includes transmitting the intermediate representation and a request for a user identification based on the intermediate representation, the remote computing system configured to perform additional steps of the multi-step identification process using the intermediate representation and the machine-learned system. The method includes receiving a user identifier for the one or more users.
-
公开(公告)号：US10887548B2
公开(公告)日：2021-01-05
申请号：US16104572
申请日：2018-08-17
Applicant: GOOGLE LLC
Inventor： Brandon Mayer , Luke Sorenson , Matthew Frazier , Wei Guan , James McGill , Zachary Yorke , Adrien Baranes , John Finley
Abstract: A method can include determining a distance of a person's face from a camera that captured an image of the person's face, determining a size of a display in communication with the camera, and scaling the image based on the determined distance of the person's face and the determined size of the display.
-
公开(公告)号：US20190058847A1
公开(公告)日：2019-02-21
申请号：US16104572
申请日：2018-08-17
Applicant: GOOGLE LLC
Inventor： Brandon Mayer , Luke Sorenson , Matthew Frazier , Wei Guan , James McGill , Zachary Yorke , Adrien Baranes , John Finley
Abstract: A method can include determining a distance of a person's face from a camera that captured an image of the person's face, determining a size of a display in communication with the camera, and scaling the image based on the determined distance of the person's face and the determined size of the display.
-
-
-
-
-
-
Search Results
7
records
(took 0.014 seconds)
