公开(公告)号：US20240305736A1

公开(公告)日：2024-09-12

申请号：US18020832

申请日：2022-05-18

Applicant: BEIJING BAIDU NETCOM SCIENCE TECHNOLOGY CO., LTD.

Inventor： Yanlong ZHANG ,  HU Weidong ,  ZHANG Qi

IPC: H04N1/60 ,  G06T5/20 ,  G06V10/56 ,  G06V10/60

CPC classification number: H04N1/6002 ,  G06T5/20 ,  G06V10/56 ,  G06V10/60

Abstract: A method of processing a video frame, an electronic device, and a storage medium, which relate to a field of an artificial intelligence technology, in particular to fields of computer vision and deep learning technologies, and may be applied to image processing, image recognition and other scenarios. A specific implementation solution includes: determining an enhancement configuration information in response to an image enhancement request for an initial video frame, wherein the enhancement configuration information includes an information related to adjusting a current attribute value of at least one image attribute of the initial video frame to a target attribute value, and adjusting, by using an image enhancement tool, the current attribute value of the at least one image attribute of the initial video frame according to the enhancement configuration information, so as to obtain a target video frame, wherein the image enhancement tool supports an encryption function.

公开(公告)号：US20230137502A1

公开(公告)日：2023-05-04

申请号：US18091704

申请日：2022-12-30

Applicant: BEIJING BAIDU NETCOM SCIENCE TECHNOLOGY CO., LTD.

Inventor： Yingyu JI ,  Yanlong ZHANG ,  Jingjing SUN

IPC: G06T5/30 ,  G06T7/70

Abstract: A method for processing a feature image includes: grouping parameters in a parameter matrix to obtain a plurality of arrays; the parameter matrix being a matrix converted and obtained from a convolutional layer in a convolutional neural network; performing thinning processing on the parameter matrix according to parameter values in the plurality of arrays to obtain a thinned parameter matrix; performing calculation by using the thinned parameter matrix and a data matrix to determine an output feature map corresponding to the convolutional layer in the case where a sparsity of the thinned parameter matrix satisfies a predetermined condition; the data matrix including a matrix converted and obtained from an input feature map inputted into the convolutional layer.