Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
METHOD FOR OBTAINING FEATURE INFORMATION OF CATEGORY DESCRIPTION, IMAGE PROCESSING METHOD, AND DEVICE
Document Type and Number:
WIPO Patent Application WO/2023/207823
Kind Code:
A1
Abstract:
A method for obtaining feature information of a category description, an image processing method, and a device. The method can be applied in the field of image processing in the field of artificial intelligence. The method comprises: obtaining feature information of at least two category descriptions corresponding to each first category in C categories; generating predicted category information of an image according to the feature information of the at least two category descriptions corresponding to each first category and feature information of the image, the predicted category of the image being comprised in the C categories; and updating, according to a first loss function, the feature information of the at least two category descriptions corresponding to each first category. The feature information of the at least two category descriptions corresponding to each category is automatically learned, and the objective of iterative updating comprises improving a similarity between the predicted category information and correct category information. Thus, it is beneficial to obtaining a category description that is matched with a recognition task, and is beneficial to improving the adaptation degree with different images in the same category.

Inventors:
LU YUNING (CN)
LIU JIANZHUANG (CN)
TIAN XINMEI (CN)
Application Number:
PCT/CN2023/089990
Publication Date:
November 02, 2023
Filing Date:
April 23, 2023
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
HUAWEI TECH CO LTD (CN)
UNIV SCIENCE & TECHNOLOGY CHINA (CN)
International Classes:
G06V10/764
Foreign References:
CN114998643A2022-09-02
CN112418256A2021-02-26
CN113065634A2021-07-02
CN111291803A2020-06-16
US10769766B12020-09-08
Attorney, Agent or Firm:
SHENPAT INTELLECTUAL PROPERTY AGENCY (CN)
Download PDF: