公开(公告)号：US20210304438A1

公开(公告)日：2021-09-30

申请号：US17346835

申请日：2021-06-14

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

Inventor： Xiaoqing Ye ,  Zhikang Zou ,  Xiao Tan ,  Hao Sun

IPC: G06T7/73 ,  G06N3/08 ,  G06K9/00

Abstract: The present disclosure provides an object pose obtaining method, and an electronic device, relates to technology fields of image processing, computer vision, and deep learning. A detailed implementation is: extracting an image block of an object from an image, and generating a local coordinate system corresponding to the image block; obtaining 2D projection key points in an image coordinate system corresponding to a plurality of 3D key points on a 3D model of the object; converting the 2D projection key points into the local coordinate system to generate corresponding 2D prediction key points; obtaining direction vectors between each pixel point in the image block and each 2D prediction key point, and obtaining a 2D target key point corresponding to each 2D predication key point based on the direction vectors; and determining a pose of the object according to the 3D key points and the 2D target key points.

公开(公告)号：US20210279934A1

公开(公告)日：2021-09-09

申请号：US17182604

申请日：2021-02-23

Applicant: Beijing Baidu Netcom Science and Technology Co., Ltd.

Inventor： Xiang LONG ,  Xin Li ,  Henan Zhang ,  Hao Sun

IPC: G06T13/40 ,  G06T19/20 ,  G06N3/04

Abstract: A method and apparatus for generating a virtual avatar are provided. The method may include: acquiring a first avatar, and determining an expression parameter of the first avatar, where the expression parameter of the first avatar including an expression parameter of at least one of five sense organs; and determining, based on the expression parameter of at least one of the five sense organs, a target virtual avatar that is associated with an attribute of the first avatar and has an expression of the first avatar.

公开(公告)号：US20210019531A1

公开(公告)日：2021-01-21

申请号：US16830895

申请日：2020-03-26

Applicant: Beijing Baidu Netcom Science and Technology Co., Ltd.

Inventor： Xiang Long ,  Dongliang He ,  Fu Li ,  Zhizhen Chi ,  Zhichao Zhou ,  Xiang Zhao ,  Ping Wang ,  Hao Sun ,  Shilei Wen ,  Errui Ding

IPC: G06K9/00 ,  G06N3/08 ,  G06K9/62

Abstract: a method and an apparatus for classifying a video are provided. The method may include: acquiring a to-be-classified video; extracting a set of multimodal features of the to-be-classified video; inputting the set of multimodal features into a post-fusion model corresponding to each modal respectively, to obtain multimodal category information of the to-be-classified video; and fusing the multimodal category information of the to-be-classified video, to obtain category information of the to-be-classified video. This embodiment improves the accuracy of video classification.

公开(公告)号：US10861133B1

公开(公告)日：2020-12-08

申请号：US16810986

申请日：2020-03-06

Applicant: Beijing Baidu Netcom Science and Technology Co., Ltd.

Inventor： Chao Li ,  Dongliang He ,  Xiao Liu ,  Yukang Ding ,  Shilei Wen ,  Errui Ding ,  Henan Zhang ,  Hao Sun

IPC: G06T3/40

Abstract: A super-resolution video reconstruction method, device, apparatus and a computer-readable storage medium are provided. The method includes: extracting a hypergraph from consecutive frames of an original video; inputting a hypergraph vector of the hypergraph into a residual convolutional neural network to obtain an output result of the residual convolutional neural network; and inputting the output result of the residual convolutional neural network into a spatial upsampling network to obtain a super-resolution frame, wherein a super-resolution video of the original video is formed by multiple super-resolution frames.

公开(公告)号：US11748895B2

公开(公告)日：2023-09-05

申请号：US17184379

申请日：2021-02-24

Applicant: Beijing Baidu Netcom Science and Technology Co., Ltd.

Inventor： Tianwei Lin ,  Xin Li ,  Fu Li ,  Dongliang He ,  Hao Sun ,  Henan Zhang

IPC: G06K9/00 ,  G06T7/246 ,  G06N3/04 ,  G06V20/40 ,  G06F18/25

CPC classification number: G06T7/246 ,  G06F18/253 ,  G06N3/04 ,  G06V20/41 ,  G06V20/46

Abstract: A method and apparatus for processing a video frame are provided. The method may include: converting, using an optical flow generated based on a previous frame and a next frame of adjacent frames in a video, a feature map of the previous frame to obtain a converted feature map; determining, based on an error of the optical flow, a weight of the converted feature map, and obtaining a fused feature map based on a weighted result of a feature of the converted feature map and a feature of a feature map of the next frame; and updating the feature map of the next frame as the fused feature map.

公开(公告)号：US11734809B2

公开(公告)日：2023-08-22

申请号：US17174002

申请日：2021-02-11

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

Inventor： Xiang Long ,  Ping Wang ,  Zhichao Zhou ,  Fu Li ,  Dongliang He ,  Hao Sun

IPC: G06T7/00 ,  G06N3/04 ,  G06N3/08

CPC classification number: G06T7/0002 ,  G06N3/04 ,  G06N3/08 ,  G06T2207/10016 ,  G06T2207/20081 ,  G06T2207/20084 ,  G06T2207/30168

Abstract: Embodiments of the present disclosure provide a method and apparatus for processing an image, and relates to the field of computer vision technology. The method may include: acquiring a value to be processed, where the value to be processed is associated with an image to be processed; and processing the value to be processed by using a quality scoring model to generate a score of the image to be processed in a target scoring domain, where the score of the image to be processed in the target scoring domain is related to an image quality of the image to be processed.

公开(公告)号：US11615140B2

公开(公告)日：2023-03-28

申请号：US17144523

申请日：2021-01-08

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

Inventor： Xiang Long ,  Dongliang He ,  Fu Li ,  Xiang Zhao ,  Tianwei Lin ,  Hao Sun ,  Shilei Wen ,  Errui Ding

IPC: G06F16/738 ,  G06V20/40 ,  G06F18/214 ,  G06F18/25

Abstract: A method includes screening, by a video-clip screening module in a video description model, a plurality of video proposal clips acquired from a video to be analyzed, to acquire a plurality of video clips suitable for description. The plural video proposal clips acquired from the video to be analyzed may be screened by the video-clip screening module to acquire the plural video clips suitable for description; and then, each video clip is described by a video-clip describing module, thus avoiding description of all the video proposal clips, only describing the screened video clips which have strong correlation with the video and are suitable for description, removing the interference of the description of the video clips which are not suitable for description in the description of the video, guaranteeing the accuracy of the final descriptions of the video clips, and improving the quality of the descriptions of the video clips.

公开(公告)号：US11915466B2

公开(公告)日：2024-02-27

申请号：US17338328

申请日：2021-06-03

Applicant: Beijing Baidu Netcom Science and Technology Co., Ltd.

Inventor： Xipeng Yang ,  Xiao Tan ,  Hao Sun ,  Hongwu Zhang

IPC: G06T7/246 ,  G06V10/44 ,  G06T7/215 ,  G06T7/73 ,  G06F18/213 ,  G06V10/25 ,  G06V30/19 ,  G06V30/24 ,  G06V10/82

CPC classification number: G06V10/454 ,  G06F18/213 ,  G06T7/215 ,  G06T7/246 ,  G06T7/74 ,  G06V10/25 ,  G06V10/82 ,  G06V30/19173 ,  G06V30/2504 ,  G06T2207/20016 ,  G06T2207/20081 ,  G06V2201/07

Abstract: Embodiments of the present disclosure disclose a method and apparatus for determining a target anchor, a device and a storage medium. The method may include: extracting a plurality of feature maps of an original image using a feature extraction network; inputting the plurality of feature maps into a feature pyramid network to perform feature fusion, to obtain a plurality of fused feature maps; and using a region proposal network to implement operations as follows: determining an initial anchor of a network head using the fused feature map, based on a size of each fused feature map, and determining an offset parameter of the initial anchor, based on a ratio of the size of the fused feature map to the original image, and generating a plurality of candidate anchors in different directions, based on the offset parameter of the initial anchor.

公开(公告)号：US11810310B2

公开(公告)日：2023-11-07

申请号：US17335647

申请日：2021-06-01

Applicant: Beijing Baidu Netcom Science and Technology Co., Ltd.

Inventor： Dongliang He ,  Henan Zhang ,  Hao Sun

IPC: G06T7/55 ,  G06N3/08 ,  G06N3/045

CPC classification number: G06T7/55 ,  G06N3/045 ,  G06N3/08 ,  G06T2207/10032 ,  G06T2207/20084 ,  G06T2207/30181

Abstract: The satellite image processing method includes: acquiring a first target satellite image; defogging the first target satellite image through a first neural network to acquire a first satellite image; and adjusting an image quality parameter of the first satellite image through a second neural network to acquire a second satellite image.

公开(公告)号：US11727676B2

公开(公告)日：2023-08-15

申请号：US17213746

申请日：2021-03-26

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

Inventor： Yingying Li ,  Xiao Tan ,  Minyue Jiang ,  Hao Sun

IPC: G06V10/82 ,  G06T3/00 ,  G06V10/77 ,  G06V10/80 ,  G06F18/213 ,  G06F18/211 ,  G06F18/25

CPC classification number: G06V10/82 ,  G06F18/211 ,  G06F18/213 ,  G06F18/253 ,  G06T3/00 ,  G06V10/7715 ,  G06V10/806

Abstract: The present disclosure provides an image processing method. An image to be classified is input into a feature extraction model to generate N dimensional features. Dimension fusion is performed on M features of the N dimensional features to obtain M dimension fusion features. The image to be classified is processed based on M dimension fusion features and remaining features of the N dimensional features other than the M features.