Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Semantic segmentation method and device based on prior structure

A semantic segmentation and segmentation technology, applied in the field of computer vision, can solve the problem of destroying image structure information

Pending Publication Date: 2022-08-09
上海人工智能创新中心
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] However, the existing pixel-by-pixel classification paradigm focuses on improving the image pixel representation, integrating context information, and finally using pixel-by-pixel classification destroys the original structural information of the image.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Semantic segmentation method and device based on prior structure
  • Semantic segmentation method and device based on prior structure
  • Semantic segmentation method and device based on prior structure

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0049] In order to make those skilled in the art better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings in the embodiments of this specification. Obviously, the described The embodiments are only some of the embodiments of the present specification, but not all of the embodiments. Based on the embodiments in this specification, all other embodiments obtained by persons of ordinary skill in the art without creative efforts shall fall within the protection scope of this specification.

[0050] In the following description, for the purpose of illustration rather than limitation, specific details such as a specific system structure and technology are set forth in order to provide a thorough understanding of the embodiments of the present application. However, it will be apparent to those skilled in the art that the pr...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention provides a semantic segmentation method and device based on a prior structure, and the method comprises the steps: obtaining a to-be-segmented image, and extracting a feature map of the to-be-segmented image; obtaining a learnable structure token which is generated randomly; inputting the feature map and the structure token into at least one layer of interaction structure, and outputting the structure token after interaction; the interactive structure is used for extracting structural features from the feature map and endowing the structural features with structural tokens; and inputting the structure token after interaction into the convolution block to obtain a segmentation image of the to-be-segmented image. According to the scheme, the structural features of the image can be reserved, and the segmentation image is directly predicted.

Description

technical field [0001] The invention belongs to the technical field of computer vision, and particularly relates to a method and device for semantic segmentation based on a priori structure. Background technique [0002] Semantic segmentation is one of the most important contents in computer vision. At present, there is only one paradigm of semantic segmentation, which is the pixel-by-pixel classification paradigm. [0003] Looking at the task of semantic segmentation from the perspective of pixel-wise classification, existing deep learning methods based on pixel-wise classification first learn the representation of each pixel through the encoder, and then classify each pixel into a specific class to obtain a semantic mask. code. [0004] However, the existing pixel-by-pixel classification paradigms focus on improving the pixel representation of images, fusing contextual information, and finally using pixel-by-pixel classification destroys the original structural informati...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06V10/26G06V10/42G06V10/80G06V10/764G06V10/82G06K9/62G06N3/04G06N3/08
CPCG06V10/26G06V10/42G06V10/806G06V10/764G06V10/82G06N3/08G06N3/045G06F18/24Y02T10/40
Inventor 林方坚梁展豪何军军郑淼田生伟陈恺
Owner 上海人工智能创新中心
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com