公开(公告)号：US11314800B2

公开(公告)日：2022-04-26

申请号：US16690387

申请日：2019-11-21

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

Inventor： Pingping Huang ,  Min Qiao ,  Ying Li ,  Jianhui Huang

IPC: G06F16/00 ,  G06F16/532 ,  G06F16/583 ,  G06F16/535 ,  G06K9/62

Abstract: Embodiments of the present disclosure provide a method, apparatus, electronic device, and computer readable storage medium for image-based data processing. The method includes: determining, in response to an inputted query for an image, an attribute associated with an object presented in the image based on a preset mapping between the object and the attribute, further includes: determining a degree of correlation between the object and the query based on the object and the attribute, and still further includes: providing a response to the query based on the degree of correlation between the object and the query. The embodiments of the present disclosure can improve the performance of the system for image-based data processing.

公开(公告)号：US11151406B2

公开(公告)日：2021-10-19

申请号：US16690432

申请日：2019-11-21

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

Inventor： Jianhui Huang ,  Pingping Huang ,  Min Qiao ,  Ying Li

IPC: G06K9/00 ,  G06K9/46 ,  G06F16/583 ,  G06F40/30 ,  G06K9/18 ,  G06K9/62

Abstract: Embodiments of the present disclosure disclose a method, apparatus, device, and readable storage medium for image-based data processing. The method comprises: acquiring an image and a to-be-processed text; extracting features of a plurality of objects in the image, and extracting a feature of the text; fusing the features of the plurality of objects into a fused feature of the image based on a matching degree between the feature of the text and a feature of each object of the plurality of objects; and processing the text based on the fused feature of the image and the feature of the text. Embodiments of the present disclosure can accurately learn an association relationship between a text and each object in an image, and improve the processing accuracy.

公开(公告)号：US20200257922A1

公开(公告)日：2020-08-13

申请号：US16690432

申请日：2019-11-21

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

Inventor： Jianhui Huang ,  Pingping Huang ,  Min Qiao ,  Ying Li

IPC: G06K9/46 ,  G06K9/62 ,  G06F16/583 ,  G06K9/18 ,  G06F17/27

Abstract: Embodiments of the present disclosure disclose a method, apparatus, device, and readable storage medium for image-based data processing. The method comprises: acquiring an image and a to-be-processed text; extracting features of a plurality of objects in the image, and extracting a feature of the text; fusing the features of the plurality of objects into a fused feature of the image based on a matching degree between the feature of the text and a feature of each object of the plurality of objects; and processing the text based on the fused feature of the image and the feature of the text. Embodiments of the present disclosure can accurately learn an association relationship between a text and each object in an image, and improve the processing accuracy.

公开(公告)号：US11620532B2

公开(公告)日：2023-04-04

申请号：US16665882

申请日：2019-10-28

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

Inventor： Jianhui Huang ,  Min Qiao ,  Zhifan Feng ,  Pingping Huang ,  Yong Zhu ,  Yajuan Lyu ,  Ying Li

IPC: G06N3/10 ,  G06N3/04 ,  G06N3/08

Abstract: Embodiments of the present disclosure relate to a method and apparatus for generating a neural network. The method includes: acquiring a target neural network, the target neural network corresponding to a preset association relationship, and being configured to use two entity vectors corresponding to two entities in a target knowledge graph as an input, to determine whether an association relationship between the two entities corresponding to the inputted two entity vectors is the preset association relationship, the target neural network comprising a relational tensor predetermined for the preset association relationship; converting the relational tensor in the target neural network into a product of a target number of relationship matrices, and generating a candidate neural network comprising the target number of converted relationship matrices; and generating a resulting neural network using the candidate neural network.

公开(公告)号：US20200242152A1

公开(公告)日：2020-07-30

申请号：US16690387

申请日：2019-11-21

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

Inventor： Pingping Huang ,  Min Qiao ,  Ying Li ,  Jianhui Huang

IPC: G06F16/532 ,  G06K9/62 ,  G06F16/583 ,  G06F16/535

Abstract: Embodiments of the present disclosure provide a method, apparatus, electronic device, and computer readable storage medium for image-based data processing. The method includes: determining, in response to an inputted query for an image, an attribute associated with an object presented in the image based on a preset mapping between the object and the attribute, further includes: determining a degree of correlation between the object and the query based on the object and the attribute, and still further includes: providing a response to the query based on the degree of correlation between the object and the query. The embodiments of the present disclosure can improve the performance of the system for image-based data processing.