公开(公告)号：US20210216776A1

公开(公告)日：2021-07-15

申请号：US17248393

申请日：2021-01-22

Applicant: Snap Inc.

Inventor： Samuel Edward Hare ,  Fedir Poliakov ,  Guohui Wang ,  Xuehan Xiong ,  Jianchao Yang ,  Linjie Yang ,  Shah Tanmay Anilkumar

IPC: G06K9/00 ,  G06T7/246 ,  G06T1/20

Abstract: A mobile device can generate real-time complex visual image effects using asynchronous processing pipeline. A first pipeline applies a complex image process, such as a neural network, to keyframes of a live image sequence. A second pipeline generates flow maps that describe feature transformations in the image sequence. The flow maps can be used to process non-keyframes on the fly. The processed keyframes and non-keyframes can be used to display a complex visual effect on the mobile device in real-time or near real-time.

公开(公告)号：US11030721B2

公开(公告)日：2021-06-08

申请号：US16392138

申请日：2019-04-23

Applicant: Snap Inc.

Inventor： Shah Tanmay Anilkumar ,  Samuel Edward Hare ,  Guohui Wang

IPC: G06T3/40 ,  G06T5/50 ,  G06T7/11 ,  G06T7/174

Abstract: Systems and methods are provided for initiating transfer of image data corresponding to at least one predetermined level of an image pyramid comprising higher resolution to a graphic processing unit (GPU) of the computing device, calculating, by the central processing unit (CPU) of the computing device, optical flow of at least one predetermined coarse level of the image pyramid, transferring, by the CPU of the computing device, the calculated optical flow of the at least one predetermined coarse level of the image pyramid to the GPU, calculating, by the GPU of the computing device, the optical flow of the at least one predetermined level of the image pyramid comprising higher resolution, and outputting, by the GPU of the computing device, the optical flow of the image data.

公开(公告)号：US10713754B1

公开(公告)日：2020-07-14

申请号：US15908461

申请日：2018-02-28

Applicant: Snap Inc.

Inventor： Guohui Wang ,  Sumant Milind Hanumante ,  Ning Xu ,  Yuncheng Li

IPC: G06K9/62 ,  G06T3/40 ,  G06N3/08 ,  G06N3/04 ,  G06T1/20 ,  G06T11/60 ,  G06N3/063

Abstract: Remote distribution of multiple neural network models to various client devices over a network can be implemented by identifying a native neural network and remotely converting the native neural network to a target neural network based on a given client device operating environment. The native neural network can be configured for execution using efficient parameters, and the target neural network can use less efficient but more precise parameters.