公开(公告)号：US20230034794A1

公开(公告)日：2023-02-02

申请号：US17878591

申请日：2022-08-01

Applicant: Snap Inc.

Inventor： Yuncheng Li ,  Zhou Ren ,  Ning Xu ,  Enxu Yan ,  Tan Yu

IPC: G06V20/64 ,  G06T7/55 ,  G06T7/33 ,  G06K9/62 ,  G06N3/08

Abstract: Systems and methods are disclosed for capturing multiple sequences of views of a three-dimensional object using a plurality of virtual cameras. The systems and methods generate aligned sequences from the multiple sequences based on an arrangement of the plurality of virtual cameras in relation to the three-dimensional object. Using a convolutional network, the systems and methods classify the three-dimensional object based on the aligned sequences and identify the three-dimensional object using the classification.

公开(公告)号：US11755910B2

公开(公告)日：2023-09-12

申请号：US17878591

申请日：2022-08-01

Applicant: Snap Inc.

Inventor： Yuncheng Li ,  Zhou Ren ,  Ning Xu ,  Enxu Yan ,  Tan Yu

IPC: G06N3/08 ,  G06T7/55 ,  G06T7/33 ,  G06V20/64 ,  G06F18/214 ,  G06F18/2431 ,  G06V10/82 ,  G06V10/44 ,  G06V20/20 ,  G06V20/40

CPC classification number: G06N3/08 ,  G06F18/214 ,  G06F18/2431 ,  G06T7/344 ,  G06T7/55 ,  G06V10/454 ,  G06V10/82 ,  G06V20/20 ,  G06V20/41 ,  G06V20/64

Abstract: Systems and methods are disclosed for capturing multiple sequences of views of a three-dimensional object using a plurality of virtual cameras. The systems and methods generate aligned sequences from the multiple sequences based on an arrangement of the plurality of virtual cameras in relation to the three-dimensional object. Using a convolutional network, the systems and methods classify the three-dimensional object based on the aligned sequences and identify the three-dimensional object using the classification.

公开(公告)号：US11580400B1

公开(公告)日：2023-02-14

申请号：US16586635

申请日：2019-09-27

Applicant: Snap Inc.

Inventor： Enxu Yan ,  Sergey Tulyakov ,  Aleksei Podkin ,  Aleksei Stoliar

IPC: G06N3/08 ,  G06N3/082 ,  G06N3/04 ,  G06F17/16 ,  G06T7/10

Abstract: A neural network pruning system can sparsely prune neural network models using an optimizer based approach that is agnostic to the model architecture being pruned. The neural network pruning system can prune by operating on the parameter vector of the full model and the gradient vector of the loss function with respect to the model parameters. The neural network pruning system can iteratively update parameters based on the gradients, while zeroing out as many parameters as possible based a preconfigured penalty.

公开(公告)号：US12254412B2

公开(公告)日：2025-03-18

申请号：US18096338

申请日：2023-01-12

Applicant: Snap Inc.

Inventor： Enxu Yan ,  Sergey Tulyakov ,  Aleksei Podkin ,  Aleksei Stoliar

IPC: G06N3/082 ,  G06F17/16 ,  G06N3/04 ,  G06T7/10

Abstract: A neural network pruning system can sparsely prune neural network models using an optimizer based approach that is agnostic to the model architecture being pruned. The neural network pruning system can prune by operating on the parameter vector of the full model and the gradient vector of the loss function with respect to the model parameters. The neural network pruning system can iteratively update parameters based on the gradients, while zeroing out as many parameters as possible based a preconfigured penalty.

公开(公告)号：US12033078B2

公开(公告)日：2024-07-09

申请号：US18230499

申请日：2023-08-04

Applicant: Snap Inc.

Inventor： Yuncheng Li ,  Zhou Ren ,  Ning Xu ,  Enxu Yan ,  Tan Yu

IPC: G06N3/08 ,  G06F18/214 ,  G06F18/2431 ,  G06T7/33 ,  G06T7/55 ,  G06V10/44 ,  G06V10/82 ,  G06V20/20 ,  G06V20/40 ,  G06V20/64

CPC classification number: G06N3/08 ,  G06F18/214 ,  G06F18/2431 ,  G06T7/344 ,  G06T7/55 ,  G06V10/454 ,  G06V10/82 ,  G06V20/20 ,  G06V20/41 ,  G06V20/64

Abstract: Systems and methods are disclosed for capturing multiple sequences of views of a three-dimensional object using a plurality of virtual cameras. The systems and methods generate aligned sequences from the multiple sequences based on an arrangement of the plurality of virtual cameras in relation to the three-dimensional object. Using a convolutional network, the systems and methods classify the three-dimensional object based on the aligned sequences and identify the three-dimensional object using the classification.

公开(公告)号：US11704893B2

公开(公告)日：2023-07-18

申请号：US17465001

申请日：2021-09-02

Applicant: Snap Inc.

Inventor： Zhou Ren ,  Yuncheng Li ,  Ning Xu ,  Enxu Yan ,  Tan Yu

IPC: G06V10/54 ,  G06V10/44 ,  G11B27/10 ,  H04N9/87 ,  G06V20/40 ,  G06V10/764 ,  G06V10/82

CPC classification number: G06V10/454 ,  G06V10/764 ,  G06V10/82 ,  G06V20/41 ,  G06V20/46 ,  G11B27/102 ,  H04N9/87

Abstract: Aspects of the present disclosure involve a system comprising a storage medium storing a program and method for receiving a video comprising a plurality of video segments; selecting a target action sequence that includes a sequence of action phases; receiving features of each of the video segments; computing, based on the received features, for each of the plurality of video segments, a plurality of action phase confidence scores indicating a likelihood that a given video segment includes a given action phase of the sequence of action phases; identifying a set of consecutive video segments of the plurality of video segments that corresponds to the target action sequence, wherein video segments in the set of consecutive video segments are arranged according to the sequence of action phases; and generating a display of the video that includes the set of consecutive video segments and skips other video segments in the video.

公开(公告)号：US11158351B1

公开(公告)日：2021-10-26

申请号：US16228120

申请日：2018-12-20

Applicant: Snap Inc.

Inventor： Zhou Ren ,  Yuncheng Li ,  Ning Xu ,  Enxu Yan ,  Tan Yu

IPC: G11B27/10 ,  H04N9/87 ,  G06K9/00

Abstract: Aspects of the present disclosure involve a system comprising a storage medium storing a program and method for receiving a video comprising a plurality of video segments; selecting a target action sequence that includes a sequence of action phases; receiving features of each of the video segments; computing, based on the received features, for each of the plurality of video segments, a plurality of action phase confidence scores indicating a likelihood that a given video segment includes a given action phase of the sequence of action phases; identifying a set of consecutive video segments of the plurality of video segments that corresponds to the target action sequence, wherein video segments in the set of consecutive video segments are arranged according to the sequence of action phases; and generating a display of the video that includes the set of consecutive video segments and skips other video segments in the video.

公开(公告)号：US20230376757A1

公开(公告)日：2023-11-23

申请号：US18230499

申请日：2023-08-04

Applicant: Snap Inc.

Inventor： Yuncheng Li ,  Zhou Ren ,  Ning Xu ,  Enxu Yan ,  Tan Yu

IPC: G06N3/08 ,  G06T7/55 ,  G06T7/33 ,  G06V20/64 ,  G06F18/214 ,  G06F18/2431 ,  G06V10/82 ,  G06V10/44 ,  G06V20/20 ,  G06V20/40

CPC classification number: G06N3/08 ,  G06T7/55 ,  G06T7/344 ,  G06V20/64 ,  G06F18/214 ,  G06F18/2431 ,  G06V10/82 ,  G06V10/454 ,  G06V20/20 ,  G06V20/41

Abstract: Systems and methods are disclosed for capturing multiple sequences of views of a three-dimensional object using a plurality of virtual cameras. The systems and methods generate aligned sequences from the multiple sequences based on an arrangement of the plurality of virtual cameras in relation to the three-dimensional object. Using a convolutional network, the systems and methods classify the three-dimensional object based on the aligned sequences and identify the three-dimensional object using the classification.

公开(公告)号：US20230146865A1

公开(公告)日：2023-05-11

申请号：US18096338

申请日：2023-01-12

Applicant: Snap Inc.

Inventor： Enxu Yan ,  Sergey Tulyakov ,  Aleksei Podkin ,  Aleksei Stoliar

IPC: G06N3/082 ,  G06F17/16 ,  G06N3/04 ,  G06T7/10

CPC classification number: G06N3/082 ,  G06F17/16 ,  G06N3/04 ,  G06T7/10 ,  G06T2207/20081 ,  G06T2207/20084

Abstract: A neural network pruning system can sparsely prune neural network models using an optimizer based approach that is agnostic to the model architecture being pruned. The neural network pruning system can prune by operating on the parameter vector of the full model and the gradient vector of the loss function with respect to the model parameters. The neural network pruning system can iteratively update parameters based on the gradients, while zeroing out as many parameters as possible based a preconfigured penalty.42188

公开(公告)号：US11410439B2

公开(公告)日：2022-08-09

申请号：US16870138

申请日：2020-05-08

Applicant: Snap Inc.

Inventor： Yuncheng Li ,  Zhou Ren ,  Ning Xu ,  Enxu Yan ,  Tan Yu

IPC: G06V20/64 ,  G06T7/55 ,  G06T7/33 ,  G06K9/62 ,  G06N3/08

Abstract: Systems and methods are disclosed for capturing multiple sequences of views of a three-dimensional object using a plurality of virtual cameras. The systems and methods generate aligned sequences from the multiple sequences based on an arrangement of the plurality of virtual cameras in relation to the three-dimensional object. Using a convolutional network, the systems and methods classify the three-dimensional object based on the aligned sequences and identify the three-dimensional object using the classification.