公开(公告)号：US20210312240A1

公开(公告)日：2021-10-07

申请号：US17348285

申请日：2021-06-15

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

Inventor： Xiaodi WANG ,  Shumin HAN ,  Yuan FENG ,  Ying XIN ,  Bin ZHANG ,  Shufei LIN ,  Pengcheng YUAN ,  Xiang LONG ,  Yan PENG ,  Honghui ZHENG

IPC: G06K9/62 ,  G06K9/46 ,  G06N3/02

Abstract: A header model for instance segmentation includes a target box branch having a first branch and a second branch, where the first branch is configured to process an inputted first feature map to obtain class information and confidence of a target box, and the second branch is configured to process the first feature map to obtain location information of the target box. The header model also includes a mask branch configured to process an inputted second feature map to obtain mask information, wherein the second feature map is a feature map outputted by an ROI extraction module, and the first feature map is a feature map resulting from a pooling performed on the second feature map.

公开(公告)号：US20220114776A1

公开(公告)日：2022-04-14

申请号：US17280142

申请日：2020-07-03

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

Inventor： Xianglong XU ,  Jianfeng ZHU ,  Jiahua CUI ,  Jing XIANG ,  Hongtao LI ,  Chen HAN ,  Shufei LIN ,  Ying SU ,  Shicao LI ,  Huiqin LI ,  Xiaochu GAN ,  Fei GAO ,  Jiale YANG ,  Xueyun MA ,  Guohong LI

IPC: G06T11/60 ,  G06F40/279 ,  G06V10/74 ,  G06F40/30 ,  G06V40/16

Abstract: Provided are an emoticon package generation method and apparatus, a device and a medium which relate to the field of graphic processing and in particular to Internet technologies. The specific implementation solution is: determining at least one of associated text of an emoticon picture or a similar emoticon package of an emoticon picture, where the associated text of the emoticon picture includes at least one of main part information, scenario information, emotion information, action information or connotation information; determining target matching text from the at least one of the associated text of the emoticon picture or associated text of the similar emoticon package; and superimposing the target matching text on the emoticon picture to generate a new emoticon package.

公开(公告)号：US20210390682A1

公开(公告)日：2021-12-16

申请号：US17116597

申请日：2020-12-09

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

Inventor： Shufei LIN ,  Jianfeng ZHU ,  Pengcheng YUAN ,  Bin ZHANG ,  Shumin HAN ,  Yingbo XU ,  Yuan FENG ,  Ying XIN ,  Xiaodi WANG ,  Jingwei LIU ,  Shilei WEN ,  Hongwu ZHANG ,  Errui DING

IPC: G06T7/00 ,  G06K9/62 ,  G06K9/46 ,  G06N3/08 ,  G06N3/04

Abstract: A method for detecting a surface defect, a method for training model, an apparatus, a device, and a medium, are provided. The method includes: inputting a surface image of the article for detection into a defect detection model to perform a defect detection, and acquiring a defect detection result output by the defect detection model; inputting a surface image of a defective article determined to be defective into an image discrimination model based on the defect detection result to determine whether the surface image of the defective article is defective, wherein the image discrimination model is a trained generative adversarial networks model, and the generative adversarial networks model is obtained by training using a surface image of a defect-free good article; and adjusting the defect detection result of the surface image of the defective article according to a determination result of the image discrimination model.

公开(公告)号：US20210350173A1

公开(公告)日：2021-11-11

申请号：US17379428

申请日：2021-07-19

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

Inventor： Xiang LONG ,  Yan PENG ,  Shufei LIN ,  Ying XIN ,  Bin ZHANG ,  Pengcheng YUAN ,  Xiaodi WANG ,  Yuan FENG ,  Shumin HAN

IPC: G06K9/62 ,  G06K9/46 ,  G02B21/36

Abstract: Provided are a method and apparatus for evaluating image relative definition, a device and a medium, relating to technologies such as computer vision, deep learning and intelligent medical. A specific implementation solution is: extracting a multi-scale feature of each image in an image set, where the multi-scale feature is used for representing definition features of objects having different sizes in an image; and scoring relative definition of each image in the image set according to the multi-scale feature by using a relative definition scoring model pre-trained, where the purpose for training the relative definition scoring model is to learn a feature related to image definition in the multi-scale feature.