Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
IMAGE PROCESSING METHOD AND APPARATUS, DEVICE, AND READABLE STORAGE MEDIUM
Document Type and Number:
WIPO Patent Application WO/2024/011835
Kind Code:
A1
Abstract:
The present application relates to the technical field of image processing, and provides an image processing method and apparatus, a device, and a readable storage medium, used for solving, to a certain extent, the problem of an inaccurate image segmentation result in existing image processing methods. The method comprises: acquiring an image to be processed; and by means of a trained image segmentation model, processing said image so as to obtain a segmented image. In the image segmentation model, M first encoding feature layers are connected to M first decoding feature layers in a one-to-one correspondence mode by means of attention mechanism modules, and N second encoding feature layers are connected to N second decoding feature layers in a one-to-one correspondence mode by means of self-attention mechanism modules. The attention mechanism modules are configured to perform feature enhancement processing on low-level features outputted by the corresponding first encoding feature layers, so as to obtain target region features; and the self-attention mechanism modules are configured to extract global context information from high-level semantic features outputted by the corresponding second encoding feature layers.

Inventors:
SI WEIXIN (CN)
LI CAIZI (CN)
Application Number:
PCT/CN2022/138163
Publication Date:
January 18, 2024
Filing Date:
December 09, 2022
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
SHENZHEN INST ADV TECH (CN)
International Classes:
G06T7/11
Domestic Patent References:
WO2022032823A12022-02-17
Foreign References:
CN115330813A2022-11-11
CN111260663A2020-06-09
CN112465828A2021-03-09
CN113744284A2021-12-03
CN114581662A2022-06-03
US20210390700A12021-12-16
Attorney, Agent or Firm:
BEIJING ZHONG XUN TONG DA INTELLECTUAL PROPERTY AGENCY CO., LTD. (CN)
Download PDF: