-
公开(公告)号:US10198820B1
公开(公告)日:2019-02-05
申请号:US15420652
申请日:2017-01-31
Applicant: Google LLC
Inventor: Peter Joseph McNerney , Nolan Andrew Miller
Abstract: Implementations generally relate to object based image editing. In some implementations, a method includes segmenting an image into object data by identifying one or more object classifications in the image and storing at least one locator for one or more regions of the image corresponding to each instance of the object classification. The method further includes receiving a selection of a representative portion of the segmented image from a user, and matching the representative portion with the object data to determine at least one matched object classification associated with the representative portion. The method further includes presenting the user with one or more of the matched object classifications for the user to instruct one or more edit operations to be applied to at least one object represented by the matched object classification.
-
公开(公告)号:US20240013772A1
公开(公告)日:2024-01-11
申请号:US18471627
申请日:2023-09-21
Applicant: Google LLC
Inventor: Nolan Andrew Miller , Ramin Mehran
Abstract: A method for multi-channel voice activity detection includes receiving a sequence of input frames characterizing streaming multi-channel audio captured by an array of microphones. Each channel of the streaming multi-channel audio includes respective audio features captured by a separate dedicated microphone. The method also includes determining, using a location fingerprint model, a location fingerprint indicating a location of a source of the multi-channel audio relative to the user device based on the respective audio features of each channel of the multi-channel audio. The method also includes generating an output from an application-specific classifier. The first score indicates a likelihood that the multi-channel audio corresponds to a particular audio type that the particular application is configured to process. The method also includes determining whether to accept or reject the multi-channel audio for processing by the particular application based on the first score generated as output from the application-specific classifier.
-
公开(公告)号:US20230316081A1
公开(公告)日:2023-10-05
申请号:US18011873
申请日:2022-05-06
Applicant: Google LLC
Inventor: Mark Sandler , Andrey Zhmoginov , Thomas Edward Madams , Maksym Vladymyrov , Nolan Andrew Miller , Blaise Aguera-Arcas , Andrew Michael Jackson
Abstract: The present disclosure provides a new type of generalized artificial neural network where neurons and synapses maintain multiple states. While classical gradient-based backpropagation in artificial neural networks can be seen as a special case of a two-state network where one state is used for activations and another for gradients with update rules derived from the chain rule, example implementations of the generalized framework proposed herein may additionally: have neither explicit notion of nor ever receive gradients; contain more than two states; and/or implement or apply learned (e.g., meta-learned) update rules that control updates to the state(s) of the neuron during forward and/or backward propagation of information.
-
公开(公告)号:US12154547B2
公开(公告)日:2024-11-26
申请号:US18471627
申请日:2023-09-21
Applicant: Google LLC
Inventor: Nolan Andrew Miller , Ramin Mehran
Abstract: A method for multi-channel voice activity detection includes receiving a sequence of input frames characterizing streaming multi-channel audio captured by an array of microphones. Each channel of the streaming multi-channel audio includes respective audio features captured by a separate dedicated microphone. The method also includes determining, using a location fingerprint model, a location fingerprint indicating a location of a source of the multi-channel audio relative to the user device based on the respective audio features of each channel of the multi-channel audio. The method also includes generating an output from an application-specific classifier. The first score indicates a likelihood that the multi-channel audio corresponds to a particular audio type that the particular application is configured to process. The method also includes determining whether to accept or reject the multi-channel audio for processing by the particular application based on the first score generated as output from the application-specific classifier.
-
公开(公告)号:US20190236786A1
公开(公告)日:2019-08-01
申请号:US16266609
申请日:2019-02-04
Applicant: Google LLC
Inventor: Peter Joseph McNerney , Nolan Andrew Miller
CPC classification number: G06T7/11 , G06K9/00221 , G06K9/00664 , G06K9/3233 , G06K9/46 , G06K9/6215 , G06K9/6267 , G06K2009/4666 , G06T5/00 , G06T11/60 , G06T2207/20012 , G06T2207/20104 , G06T2207/30201
Abstract: Implementations generally relate to object based image editing. In some implementations, a method includes segmenting an image into object data by identifying one or more object classifications in the image and storing at least one locator for one or more regions of the image corresponding to each instance of the object classification. The method further includes receiving a selection of a representative portion of the segmented image from a user, and matching the representative portion with the object data to determine at least one matched object classification associated with the representative portion. The method further includes presenting the user with one or more of the matched object classifications for the user to instruct one or more edit operations to be applied to at least one object represented by the matched object classification.
-
公开(公告)号:US11380302B2
公开(公告)日:2022-07-05
申请号:US17077679
申请日:2020-10-22
Applicant: Google LLC
Inventor: Nolan Andrew Miller , Ramin Mehran
Abstract: A method for multi-channel voice activity detection includes receiving a sequence of input frames characterizing streaming multi-channel audio captured by an array of microphones. Each channel of the streaming multi-channel audio includes respective audio features captured by a separate dedicated microphone. The method also includes determining, using a location fingerprint model, a location fingerprint indicating a location of a source of the multi-channel audio relative to the user device based on the respective audio features of each channel of the multi-channel audio. The method also includes generating an output from an application-specific classifier. The first score indicates a likelihood that the multi-channel audio corresponds to a particular audio type that the particular application is configured to process. The method also includes determining whether to accept or reject the multi-channel audio for processing by the particular application based on the first score generated as output from the application-specific classifier.
-
公开(公告)号:US11790888B2
公开(公告)日:2023-10-17
申请号:US17806198
申请日:2022-06-09
Applicant: Google LLC
Inventor: Nolan Andrew Miller , Ramin Mehran
Abstract: A method for multi-channel voice activity detection includes receiving a sequence of input frames characterizing streaming multi-channel audio captured by an array of microphones. Each channel of the streaming multi-channel audio includes respective audio features captured by a separate dedicated microphone. The method also includes determining, using a location fingerprint model, a location fingerprint indicating a location of a source of the multi-channel audio relative to the user device based on the respective audio features of each channel of the multi-channel audio. The method also includes generating an output from an application-specific classifier. The first score indicates a likelihood that the multi-channel audio corresponds to a particular audio type that the particular application is configured to process. The method also includes determining whether to accept or reject the multi-channel audio for processing by the particular application based on the first score generated as output from the application-specific classifier.
-
公开(公告)号:US20230210457A1
公开(公告)日:2023-07-06
申请号:US18087012
申请日:2022-12-22
Applicant: Google LLC
Inventor: Nolan Andrew Miller
IPC: A61B5/00 , A61B5/11 , A61B5/1455 , A61B5/08 , A61B5/103
CPC classification number: A61B5/4866 , A61B5/1118 , A61B5/14552 , A61B5/7257 , A61B5/0816 , A61B5/742 , A61B5/1032 , A61B2562/0219
Abstract: A method for estimating a metabolic rate of a user includes obtaining pulse oximetry data for the user for a period of time. The method includes determining a rate of decline in oxygen saturation of blood of the user that is associated with a breathing rate of the user for the period of time based, at least in part, on the pulse oximetry data. The method includes estimating the metabolic rate of the user for the period of time based, at least in part, on the rate of decline in the oxygen saturation of the blood that is associated with the breathing rate of the user. The method includes providing a notification indicative of the metabolic rate for the period of time.
-
公开(公告)号:US20220310060A1
公开(公告)日:2022-09-29
申请号:US17806198
申请日:2022-06-09
Applicant: Google LLC
Inventor: Nolan Andrew Miller , Ramin Mehran
Abstract: A method for multi-channel voice activity detection includes receiving a sequence of input frames characterizing streaming multi-channel audio captured by an array of microphones. Each channel of the streaming multi-channel audio includes respective audio features captured by a separate dedicated microphone. The method also includes determining, using a location fingerprint model, a location fingerprint indicating a location of a source of the multi-channel audio relative to the user device based on the respective audio features of each channel of the multi-channel audio. The method also includes generating an output from an application-specific classifier. The first score indicates a likelihood that the multi-channel audio corresponds to a particular audio type that the particular application is configured to process. The method also includes determining whether to accept or reject the multi-channel audio for processing by the particular application based on the first score generated as output from the application-specific classifier.
-
公开(公告)号:US20220130375A1
公开(公告)日:2022-04-28
申请号:US17077679
申请日:2020-10-22
Applicant: Google LLC
Inventor: Nolan Andrew Miller , Ramin Mehran
Abstract: A method for multi-channel voice activity detection includes receiving a sequence of input frames characterizing streaming multi-channel audio captured by an array of microphones. Each channel of the streaming multi-channel audio includes respective audio features captured by a separate dedicated microphone. The method also includes determining, using a location fingerprint model, a location fingerprint indicating a location of a source of the multi-channel audio relative to the user device based on the respective audio features of each channel of the multi-channel audio. The method also includes generating an output from an application-specific classifier. The first score indicates a likelihood that the multi-channel audio corresponds to a particular audio type that the particular application is configured to process. The method also includes determining whether to accept or reject the multi-channel audio for processing by the particular application based on the first score generated as output from the application-specific classifier.
-
-
-
-
-
-
-
-
-